Mycobacteriophage-mediated lysis: the role of LysB proteins · 2019-03-10 · x Doutora Matilde da Luz dos Santos Duque da Fonseca e Castro, Professora Catedrática da Faculdade de

UNIVERSIDADE DE LISBOA

FACULDADE DE FARMÁCIA

Mycobacteriophage-mediated lysis:

the role of LysB proteins

DOUTORAMENTO EM FARMÁCIA

MICROBIOLOGIA

Adriano Marcelo Dos Santos Gigante

Tese orientada pela Professora Doutora Madalena Maria Vilela Pimentel, coorientada pelo Professor Doutor José António Frazão Moniz-Pereira, especialmente elaborada

para a obtenção do grau de doutor.

2018

UNIVERSIDADE DE LISBOA

FACULDADE DE FARMÁCIA

Mycobacteriophage-mediated lysis: the role of LysB proteins

Adriano Marcelo Dos Santos Gigante

Orientadores: Professora Doutora Madalena Maria Vilela Pimentel

Professor Doutor José António Frazão Moniz-Pereira

Tese especialmente elaborada para a obtenção do grau de Doutor em Farmácia, especialidade em Microbiologia.

Júri:

Presidente:

• Doutora Matilde da Luz dos Santos Duque da Fonseca e Castro, Professora Catedrática da Faculdade de Farmácia da Universidade de Lisboa

Vogais:

• Doutor Sérgio Joaquim Raposo Filipe, Professor Auxiliar da Faculdade de Ciências e Tecnologia da Universidade Nova de Lisboa;

• Doutora Joana Cecília Valente Rodrigues Azeredo, Professora Auxiliar com Agregação da Escola de Engenharia da Universidade do Minho;

• Doutora Madalena Maria Vilela Pimentel, Professora Auxiliar da Faculdade de Farmácia da Universidade de Lisboa, Orientadora

• Doutora Ana Paula Costa dos Santos Peralta, Professora Auxiliar da Faculdade de Farmácia da Universidade de Lisboa

• Doutor Carlos Jorge Sousa de São-José, Professor Auxiliar da Faculdade de Farmácia da Universidade de Lisboa

Trabalho financiado pela Fundação para a Ciência e tecnologia (FCT) através da Bolsa individual de doutoramento SFRH/BD/87685/2012.

2018

O autor dos trabalhos descritos nesta dissertação, Adriano Marcelo Dos Santos Gigante,

foi apoiado pela Fundação para a Ciência e Tecnologia com concessão da bolsa de

doutoramento SFRH/BD/87685/2012.

O trabalho apresentado nesta dissertação foi maioritariamente desenvolvido no Centro de

Patogénese Molecular ̶ Unidade de Retrovírus e Infeções Associadas (CPM-URIA) e no

instituto de Investigação do Medicamento (iMedULisboa) da Faculdade de Farmácia da

Universidade de Lisboa (FFUL), sob orientação da Professora Doutora Madalena Maria

Vilela Pimentel e coorientação do Professor Doutor José António Frazão Moniz-Pereira.

THESIS OUTPUT

The studies presented in this thesis are also the subject of the following scientific

papers or congress poster/oral communications:

Gigante, A. M., Hampton, C. M., Dillard, R. S., Gil, F., Catalão, M. J., Moniz-Pereira, J.,

Wright, E. R., Pimentel, M., (2017) The Ms6 Mycolyl-Arabinogalactan Esterase LysB is

Essential for an Efficient Mycobacteriophage-Induced Lysis. Viruses., 9, 343.

Gigante, A.M., Catalão, M.J., Olivença, F., Moniz-Pereira, J., Filipe, S., Pimentel, M.,

The N-terminal of mycobacteriophage Ms6 LysB reveals peptidoglycan binding capacity.

Manuscript in preparation.

Gigante, A.M., Gil F., Moniz-Pereira, J., Leandro, P., Pimentel, M., Comparative

modelling and enzymatic activity of LysB mycolylarabinogalactan esterases from Ms6,

Adjutor, Trixie and U2 mycobacteriophages. Manuscript in preparation.

Oral communications in conferences

A.M. Gigante, F. Martins, F. Gil, M. Pimentel. The impact of lysB in mycobacteria phage-

mediated lysis. 2013, MICROBIOTEC ’13, Aveiro (Portugal).

A.M. Gigante, C. Hampton, R. Dillard, J. Moniz-Pereira, E. Wright, M. Pimentel.

Mycobacteriophage Ms6 LysB: role on the mycobacteria lysis process. 2016,

Bacteriophages: Theoretical and Practical Aspects of Their Application in Medicine,

Veterinary and Food, Moscow (Russia).


Assessing the role of mycobacteriophage Ms6 LysB in phage-mediated host lysis. 2017,

9th iMed.ULisboa Post-graduate Students Meeting, Lisbon (Portugal).

Posters in conferences

A.M. Gigante, F. Martins, F. Gil, J. Moniz-Pereira, M. Pimentel. The Ms6 LysB

significance in mycobacteria phage-mediated lysis. 2014, 6th iMed.ULisboa Post-

graduate Students Meeting, Lisbon (Portugal).

A.M. Gigante, F. Martins, F. Gil, M. Pimentel. The impact of Ms6 LysB in mycobacteria

phage-mediated lysis. 2014, Virus Of Microbes, Zürich (Switzerland).

A.M. Gigante, C. Hampton, R. Dillard, J. Moniz-Pereira, E. Wright, M. Pimentel. Role

of Ms6 LysB on host cell lysis assessed by Cryo-Electron Microscopy. 2015, 7th

iMed.ULisboa Post-graduate Students Meeting, Lisbon (Portugal).


Mycobacteriophage Ms6: role of LysB on host cell lysis assessed by Cryo-Electron

Microscopy. 2015, MICROBIOTEC ’15, Évora (Portugal).

A.M. Gigante, M.J. Catalão, J. Moniz-Pereira, S. Filipe, M. Pimentel. The N-terminus of

mycobacteriophage Ms6 LysB has peptidoglycan binding capacity. 2016, 8th

iMed.ULisboa Post-graduate Students Meeting, Lisbon (Portugal).


Mycobacteriophage Ms6 LysB: an essential lysis protein for an efficient lysis. 2016,

Molecular Genetics of Bacteria and Phages Meeting, Madison (U.S.A.).

De acordo com o ponto 1 nº45 do regulamento de Estudos Pós-graduados da Universidade

de Lisboa, deliberação nº 4624/2012, publicado em Diário da República – II Série nº 65

– 30 de Março de 2012, o autor desta dissertação declara que participou na conceção e

execução do trabalho experimental, interpretação dos resultados obtidos e redação dos

manuscritos.

i

TABLE OF CONTENTS

AGRADECIMENTOS ................................................................................................. vii

RESUMO ........................................................................................................................ xi

ABSTRACT .................................................................................................................. xv

ABREVIATIONS ....................................................................................................... xvii

CHAPTER 1.

GENERAL INTRODUCTION

1.1 The bacteriophage .................................................................................................... 3

1.1.1 Phage discovery and ecology ............................................................................... 3

1.1.2 Phage Classification and Characterization ........................................................... 4

1.1.3 Phage life cycles .................................................................................................. 5

1.1.4 Mycobacteriophages, the viruses of mycobacteria .............................................. 9

1.2 Cell barriers to phage release ................................................................................ 10

1.2.1 The cytoplasmic membrane ............................................................................... 12

1.2.2 The cell wall - General considerations .............................................................. 12

1.2.2.1 Peptidoglycan: the glycan strands and peptic cross-linking ....................... 13

1.2.3 Gram-positive cell envelope .............................................................................. 15

1.2.4 Gram-negative cell envelope ............................................................................. 16

1.2.5 Mycobacterial cell envelope .............................................................................. 17

1.2.5.1 Global structure .......................................................................................... 17

1.2.5.2 Peptidoglycan modifications ...................................................................... 20

1.2.5.3 Arabinogalactan .......................................................................................... 20

1.2.5.4 Lipoarabinomannan .................................................................................... 21

1.2.5.5 Trehalose dimycolate and other extractable lipids ..................................... 21

1.3 Overview of phage lysis players ............................................................................. 23

1.3.1 Holins: a molecular hourglass of lysis ............................................................... 24

1.3.2 Antiholin: a post-translational lysis regulator .................................................... 25

ii

1.3.3 Endolysins: phage-encoded enzymes to degrade the peptidoglycan ................. 27

1.3.4 Spanins: proteins for membrane fusion ............................................................. 29

1.4 Phage-mediated host cell lysis ................................................................................ 32

1.4.1 The λ phage as the lysis model paradigm .......................................................... 32

1.4.2 Holin-independent export of endolysins ............................................................ 36

1.4.3 fOg44, holin-independent export of Lys44 SP-endolysin ................................. 36

1.4.4 SV1 holin-independent endolysin export .......................................................... 37

1.4.5 Holin-independent export of SAR-endolysins in coliphage systems ................ 38

1.4.6 φKMV holin-independent SAR-endolysin export ............................................. 39

1.4.7 ERA103, holin-independent export of Lys103 SAR-endolysin .......................... 39

1.5 The mycobacteriophage Ms6 ................................................................................. 40

1.1.2. Genetic organization of the lysis module ......................................................... 42

1.5.1 Ms6 LysB is a mycolyl-arabinogalactan esterase .............................................. 44

1.5.2 Ms6 current lysis model ..................................................................................... 46

1.6 Thesis goals .............................................................................................................. 48

1.7 References ................................................................................................................ 49

CHAPTER 2.

THE MS6 MYCOLYL-ARABINOGALACTAN ESTERASE LYSB IS

ESSENTIAL FOR AN EFFICIENT MYCOBACTERIOPHAGE INDUCED LYSIS

Acknowledgements…………………………………………………..…..…………....71

Abstract……………………………………………………………….….……………73

2.1. Introduction ............................................................................................................ 75

2.2. Results .................................................................................................................. 78

2.2.1 Ms6 lysB deletion decreases viral progeny release ............................................ 78

2.2.2 Ms6 is trapped in cell debris in absence of LysB .............................................. 80

2.2.3 Cryo-EM shows incomplete cell lysis in absence of Ms6 LysB ....................... 81

iii

2.3. Discussion ............................................................................................................... 84

2.4. Materials and methods .......................................................................................... 88

2.4.1 Bacterial strains, phages, plasmids and culture conditions ................................ 88

2.4.2 Construction of Ms6 mutant phage.................................................................... 89

2.4.3 Plasmid construction .......................................................................................... 90

2.4.4 One step growth and single burst experiments .................................................. 90

2.4.5 Determination of the number of phage particles released during the infection

cycle .............................................................................................................. 91

2.4.6 Cryo-transmission electron microscopy sample preparation, imaging and image

processing .............................................................................................................. 91

2.4.7 Nucleotide sequence accession numbers ........................................................... 92

2.5. References ............................................................................................................... 92

2.6. Supplementary materials ...................................................................................... 97

CHAPTER 3.

THE N-TERMINAL OF MYCOBACTERIOPHAGE MS6 LYSB REVEALS

PEPTIDOGLYCAN BINDING CAPACITY

Acknowledgements…………………………………………………...…..……….....103

Abstract……………………………………………………………………….……...105

3.1. Introduction .......................................................................................................... 107

3.2. Results ................................................................................................................ 109

3.2.1 Sequence analysis of Ms6 LysB reveals a putative PGBD ............................. 109

3.2.2 The Ms6 LysB N- terminus binds to mycobacterial cells ............................... 111

3.2.3 Ms6 LysB N- terminus binds to peptidoglycan ............................................... 114

3.2.4 Deletion of PGBD from LysB results in a faster rise period ........................... 118

3.3. Discussion ............................................................................................................. 120

3.4. Materials and Methods........................................................................................ 123

3.4.1 Bacterial strains, plasmids, bacteriophages and culture conditions ................. 123

iv

3.4.2 General DNA techniques and cloning procedures ........................................... 123

3.4.3 Plasmid construction ........................................................................................ 123

3.4.4 Protein expression and purification ................................................................. 125

3.4.5 Cells binding and fluorescence assay .............................................................. 126

3.4.6 Preparation and purification of peptidoglycan ................................................ 126

3.4.7 Peptidoglycan binding assays .......................................................................... 127

3.4.8 Construction of Ms6lysB∆PGBD .................................................................... 128

3.4.9 One step growth experiments .......................................................................... 128

3.5. References ............................................................................................................. 129

3.6. Supplementary material ...................................................................................... 133

CHAPTER 4.

COMPARATIVE MODELLING AND ENZYMATIC ACTIVITY OF LYSB

MYCOLYLARABINOGALACTAN ESTERASES FROM MS6, ADJUTOR,

TRIXIE AND U2 MYCOBACTERIOPHAGES

Acknowledgements…………………………………………………………….…….141

Abstract……………………………………………………………………….…..….143

4.1. Introduction .......................................................................................................... 145

4.2. Results ................................................................................................................ 147

4.2.1 Analysis of LysB sequences alignment from different bacteriophages shows the

GXSXG conserved motif ......................................................................................... 147

4.2.2 Comparative modeling using D29 as the template allowed prediction of 3D

structure of Ms6, Trixie and U2 LysB ...................................................................... 150

4.2.3 Relative activity of Ms6, Adjutor, Trixie and U2 LysB .................................. 155

4.2.4 Ms6 LysB catalytic triad is formed by Ser168, Asp249 and His318 residues 157

4.2.5 Loss of enzymatic activity of Ala substitutions on Ms6 LysB catalytic triad

residues is corroborated by structural modeling predictions. ................................... 159

4.3. Discussion ............................................................................................................. 162

v

4.4. Material and methods.......................................................................................... 164

4.4.1 Bacterial strains, bacteriophages and plasmids ............................................... 164

4.4.2 DNA amplification and cloning ....................................................................... 165

4.4.3 Site-directed mutagenesis ................................................................................ 166

4.4.4 Expression and purification of His6-LysB proteins ......................................... 166

4.4.5 Catalytic activity assays ................................................................................... 167

4.4.6 In silico analysis .............................................................................................. 167

4.5. References ............................................................................................................. 168

4.6. Supplementary material ...................................................................................... 171

CHAPTER 5.

CONCLUDING REMARKS

5.1. Concluding Remarks ........................................................................................... 177

5.2. References ............................................................................................................. 183

vii

AGRADECIMENTOS

Este manuscrito é a prova documental de uma longa jornada durante o meu

doutoramento, a todos e tantos que me deram coragem para seguir, caminharam comigo

ou me orientaram na viagem quero deixar aqui o meu agradecimento.

Ao Professor Doutor José Moniz-Pereira o meu agradecimento é particularmente

extenso. Em 2008 enquanto aluno do MICF na FFUL foi o meu professor na cadeira de

bacteriologia laboratorial, identifico esse momento como o ponto de partida para o rumo

que tomei desde então, posso dizer com convicção que foi aí que nasceu o “bichinho”.

Depois disso foi meu orientador na cadeira de Projecto I, foi quem me apresentou nessa

altura à professora doutora Madalena Pimentel que posteriormente me orientou na cadeira

de Projecto III. Pelos motivos citados, ainda por ter sido meu co-orientador e enquanto

líder de departamento, ao Professor Doutor José Moniz-Pereira agradeço por tudo o que

me permitiu fazer, aprender e crescer durante todo este percurso, a sua contribuição foi

sem dúvida essencial. Obrigado!

À minha orientadora Professora Doutora Madalena Pimentel agradeço primeiro por

tudo o que fiz de inteligível e que trago nesta tese. O seu papel foi essencial e todos os

merecidos elogios ficam aquém do verdadeiro impacto que teve na minha vida, pela

amizade, carinho, compreensão, sinceridade, lucidez e motivação!!! Por ter acreditado

neste “saloio de gema”, mas com uma vontade bem clara de aprender com os melhores.

Obrigado!

Ao Professor Doutor Carlos São-José agradeço pelo AKTAPrime! Mas não só! A

si fiz muitas perguntas e as suas respostas sempre em mim deixaram um sentimento de

admiração. Pelos empréstimos de material, técnicas e ideias, mas também pela

disponibilidade, pensamento científico e olhar crítico só tenho, e muito, que lhe

agradecer. Aos elementos do respetivo grupo, Catarina, Sofia, Hugo e Nuno agradeço a

companhia e amizade, pelos bons momentos que passamos quando estiveram “acolhidos”

no CPM e pelas discussões construtivas nos LabMeetings de sexta-feira. Obrigado!

viii

Aos membros do antigamente designado Grupo de Microbiologia, aos Professores,

mas não só, para todas aquelas pessoas se foram cruzando com esta a minha jornada, e

todas elas tinham algo a acrescentar, aquele “protocolo que funciona mesmo”, o “truque

para isto dar bem é” ou o “eu acho que devias fazer”, a todos os que de algum modo

contribuíram deixo aqui um maiúsculo OBRIGADO!

I am sincerely thankful to Professor Elizabeth Wright for the opportunity to learn

and work in her lab. I will never forget how warmly I was welcomed as a visitor, a true

display of southern hospitality. I also want to extend the greetings to all the members of

“The Wright Lab” that I had the pleasure to learn from, Doctor Cheri Hampton, Rebecca

Dillard and Zunlong Ke, and also the members of the Robert P. Apkarian EM Core for

the invaluable technical support.

Agradeço ao Professor Doutor Sérgio Filipe pelo apoio prestado que foi sem dúvida

indispensável para a realização de todos os trabalhos de microscopia de fluorescência

apresentados nesta tese. Aos elementos do grupo do ITQB, Maria João Catalão, Gonçalo

Covas e Rita Guedes que me ajudaram todos e muito! Com vocês aprendi a “brilhar no

escuro” e passámos bons momentos. Obrigado!

Agradeço à Professora Doutora Paula Leandro pela inestimável ajuda e

ensinamentos que me deu, essencialmente na elaboração dos modelos das proteínas

apresentados nesta tese.

Chegou agora a secção dos agradecimentos mais complicada de gerir, e deixo aqui

uma linha sobre “conflito de interesses”, já sei que me vão “criticar” porque é que

menciono primeiro um ou outro, e como sei que não há disso salvação possível a ordem

apresentada é meramente aleatória, ou então cronológica ou alfabética quando possível.

Dito isto…

Ao Pedro Rodrigues, Diana Gaspar, Gisela Santos e Tânia Genebra, meus colegas

de cruzada, juntos trilhámos as mesmas pedras, ou pelo menos uns pedregulhos parecidos,

cada um de vocês é para mim especial e nós os cinco juntos somos uma “especialidade”,

cada um com o seu poder somos tipo power-ranger mas sem a parte da licra. É melhor

parar por aqui para não descarrilar a conversa (como de costume). Adoro-vos a todos!

#bestipSCever, #semfiltro. #Obrigado!

ix

Aos meus colegas de bancada, Francisco Martins, Sofia Pombo, Vanessa

Rodrigues, Nuno Bustorff, Francisco Olivença, e os “quase de bancada” da porta ao lado

Renato Alves e Sofia Romano. Todos nós estamos unidos pelas memórias, muitas, boas

e maioritariamente divertidas que o meu sorriso ao escrever estas palavras é disso prova.

Desde o eterno dilema M80vsOrbital até aos momentos de ajuda (que na micro duas mãos

nem sempre chegam), a vossa contribuição foi importantíssima. Alguns de vocês ensinei,

com todos aprendi e é também a todos que agradeço. Obrigado!

Aos membros da iMed Postgraduate Students Comission (ipSC), da qual tive o

prazer de fazer parte durante três anos, passámos tantos dias (entenda-se noites) a

organizar os meetings anuais do iMed, formatar livros de abstracts, encontrar e negociar

patrocínios, coffee-breaks e tantas, mas tantas horas de trabalho, que o sendo, foram

também momentos de aprendizagem coletiva, entreajuda e crescimento pessoal. Foi para

mim a materialização do “ser Português”, tudo se de “desenrasca”, tudo tem um “jeitinho”

e como dos fracos não reza a história, nós juntos fomos mais fortes, e é por isso que vos

agradeço. Obrigado!

Aos demais amigos, não no sentido de serem demasiados, mas porque ainda não

foram mencionados. São muitos e o espaço que tenho é pouco, mas no meu coração está

escrito o que nas linhas desta folha não cabe! Recordo que a ordem da menção é aleatória:

Catarina Pereira, Maria Pereira, Glenn Pedreiro, Sérgio Gomes, Liliana Marques, Mário

Rosa, Marisa Miguel, Filipa Maurício, Zé Fragoso, Joana Leitão, Filipa Pereira, Luís

Cruz, Vânia Guerreiro, Ricardo Santos. Como dizia um cartaz “os amigos da faculdade

são para sempre”, pelo menos onze anos já lá vão, mas mais virão e como contei sempre

com o vosso apoio sei que posso continuar a contar, por isso a todos agradeço. Obrigado!

À minha família, o que aqui escrevo é apenas uma pequena parte do muito que vos

amo! Agradeço pelo apoio incondicional e toda a ajuda que me deram para chegar a este

ponto da minha vida, só posso estar grato pelo suporte que me permitiu chegar até aqui,

e me permitirá chegar sempre onde quer que seja. Aos meus pais pela preocupação

constante, aos meus irmãos pela força e à pequena Lara, que sem saber, foi a minha maior

fonte de alegrias durante estes seus já dois anos de vida! É à minha família que dediquei

todas as minhas conquistas até aqui e dedicarei daqui em diante, é por isso óbvio que é à

minha família também que dedico esta tese.

xi

RESUMO

A maior parte dos bacteriófagos, comummente designados apenas fagos, isolados

na natureza caracterizam-se por possuírem um ADN de cadeia dupla (dsDNA) envolvido

por uma cápside proteica e por apresentarem uma cauda. Estes fagos são capazes de

efectuar um ciclo replicativo que termina com a lise da bactéria hospedeira. Para o efeito

sintetizam pelo menos duas proteínas essenciais, as holinas e as endolisinas. As holinas

são pequenas proteínas membranares que se acumulam na membrana citoplasmáitica e

que a um determinado momento formam lesões na membrana, determinando o início da

lise. A dimensão das lesões formadas pode ser desde pequenos poros, que permitem a

passagem de iões entre o espaço periplasmático e o citoplasma, dissipando o potencial de

membrana, ou grandes buracos na membrana de dimensão suficiente para permitir a

passagem das endolisinas. As endolisinas são enzimas, que cortam ligações especificas

no peptidoglicano, o principal componente da parede celular e cuja hidrólise é essencial

para lisar a célula hospedeira.

Para além do par holina-endolisina, essencial para a lise, a maioria dos fagos que

infetam bactérias Gram negativas sintetizam uma terceira classe de proteínas envolvidas

na lise, as spaninas, cuja função consiste em remover a membrana externa da célula

hospedeira garantindo uma libertação eficaz da progenia viral. O exemplo paradigmático

é o fago λ em que os genes Rz/Rz1 codificam para as spaninas Rz e Rz1, as quais

interagem entre si formando um complexo que após a degradação do peptidoglicano pela

endolisina, sofre alterações conformacionais que conduzem à fusão da membrana

citoplasmática com a membrana externa, consequentemente permitindo a libertação das

partículas fágicas para o meio extracelular envolvente. Spaninas e proteínas “spanin-

like” (proteínas cuja função se pensa ser idêntica às spaninas do fago λ) foram

posteriormente identificadas em mais alguns fagos que infetam outras bactérias Gram

negativas, por isso pensa-se que estas sejam um factor que confere vantagem para os

fagos do ponto de vista da ecologia e evolução.

Tal como as bactérias Gram-negativas as micobactérias também apresentam uma

membrana externa, embora com uma composição bastante diferente. Esta membrana é

extremamente rica em lípidos e ancora no seu folheto interno, os ácidos micólicos que

xii

ligam esta estrutura à parede celular. Para ultrapassar esta barreira os micobacteriófagos,

os fagos que infetam micobactérias, sintetizam uma proteína designada LysB. Esta

enzima é uma micolil-arabinogalactano esterase que hidrolisa a ligação éster entre os

ácidos micólicos e o arabinogalactano, desacoplando a membrana externa do PG. Neste

trabalho estudou-se o papel da proteína LysB na lise das micobactérias, utilizando como

principal modelo o micobacteriófago Ms6 que infecta Mycobacterium smegmatis.

A primeira parte do trabalho experimental apresentado nesta tese demonstra que a

proteína LysB do fago Ms6 é essencial para que a lise do hospedeiro seja eficiente. A

construção de um fago mutante (Ms6ΔlysB) no qual foi delecionado o gene lysB permitiu

estudar o fenótipo de lise por comparação com o fago selvagem (Ms6wt). O estudo dos

parâmetros biológicos permitiu mostrar que no final de uma infeção com o fago mutante

Ms6ΔlysB, o número de partículas fágicas libertadas por célula é substancialmente menor

que numa infeção com o Ms6wt, 53±14 vs 147±27 respetivamente. A quantificação destas

partículas no pellet e sobrenadante de células infectadas mostrou que a redução do número

fagos libertados resulta não de uma redução na síntese mas sim numa deficiente libertação

das partículas fágicas. 180 min após a adsorção 47% dos fagos Ms6ΔlysB ficaram retidos

no pellet contra 7% observados numa infeção com Ms6wt, revelando claramente uma

deficiência na eficácia da lise. A visualização de células de M. smegmatis infetadas com

o fago Ms6ΔlysB, utilizando crio-microscopia e crio-tomografia eletrónicas, demonstra

inequivocamente que na ausência de LysB, a lise não é eficaz pois apesar de a camada de

peptidoglicano ser destruída, a membrana externa consegue manter aprisionada parte da

progenia viral no interior da micobactéria. O efeito observado revela que a LysB tem um

papel essencial para que a lise do hospedeiro seja eficiente e na libertação da progenia

viral.

No presente estudo é demonstrado que a LysB do fago Ms6 tem uma arquitetura

modular: os primeiros 90 resíduos da região N-terminal tem semelhança estrutural com a

região N-terminal da endolisina do fago φKZ, a qual apresenta um domínio de ligação ao

peptidoglicano; a região C-terminal que contém o domínio catalítico revelou homologia

estrutural com a proteína LysB do micobacteriófago D29.

A homologia observada com a endolisina do fago φKZ serviu de base para os

estudos apresentados no capítulo 3. Através da construção de proteínas de fusão contendo

xiii

a região N-terminal fundida à proteína fluorescente enhanced green fluorescente protein

(EGFP) foi possível mostrar por microscopia de fluorescência, a capacidade de ligação a

células de Mycobacterium smegmatis, Mycobacterium vaccae, Mycobacterium bovis

BCG e Mycobacterium tuberculosis H37Ra. A ligação do peptidoglycan binding domain

(PGBD) à superfície bacteriana ocorre quando as bactérias são previamente lavadas com

SDS ou incubadas com LysB do fago Ms6, por forma a permitir a desorganização da

membrana externa e a exposição do PG ao meio envolvente. Em micobactérias não

tratadas não se verifica qualquer fluorescência, o que comprova que o alvo da ligação da

proteína de fusão não está exposto à superfície da célula. Em ensaios de “pulldown”

usando PG purificado de M. smegmatis, Pseudomonas aeruginosa, Escherichia.coli,

Bacillus.subtilis, Streptococcus pneumoniae e Staphylococcus aureus foi possível

demonstrar a capacidade de ligação da região N-terminal a PG pertencentes ao quimiotipo

A1γ observado maior afinidade para PG de M. smegmatis. Os resultados aqui

apresentados demonstram pela primeira vez, de forma experimental, que a região N-

terminal da LysB do fago Ms6 tem uma função de ligação ao peptidoglicano. Este é um

resultado importante dado que uma grande parte das proteínas LysB codificadas por

outros micobacteriófagos apresenta uma região N-terminal com semelhança ao PGBD da

LysB do Ms6. O impacto que este domínio tem na função da proteína LysB foi avaliado

através da construção de um fago mutante que produz uma variante da proteína LysB sem

a região N-terminal. O estudo do fenótipo de lise desse fago sugere que não há alterações

no tempo de lise, porém a libertação das partículas fágicas é mais rápida, não sendo clara

ainda a importância deste domínio para a lise mediada por micobacteriófagos.

A terceira parte do trabalho consistiu no estudo in silico de quatro proteínas LysB,

originadas dos fagos Ms6 Adjutor, Trixie e U2. Utilizando a única informação

cristalográfica disponível até à data para uma LysB (fago D29; PDB ID 3HC7),

obtiveram-se modelos estruturais para as LysB de Ms6, Adjutor, Trixie e U2. A avaliação

dos parâmetros de qualidade das estruturas obtidas indicou que, à exceção da LysB do

fago Adjutor, os modelos obtidos deverão refletir as estruturas das proteínas e a sua

capacidade para apresentar atividade catalítica como esterases. Na realidade a

superimposição dos domínios e resíduos envolvidos na função biológica da LysB do fago

D29 foram identificados nas estruturas geradas para as LysB de Ms6, Trixie e U2,

nomeadamente o motivo GYSQG, a tríade catalítica Ser-Asp-His e a região conservada

xiv

GNP. Para validar estas observações foram determinadas as atividades enzimáticas das

LysB dos fagos Ms6, Adjutor, Trixie e U2 utilizando como substrato ésteres de ácidos

gordos de diferentes cadeias hidrocarbonadas (C4, C12 e C14) e p-nitrofenol. Como

esperado, todas as proteínas estudadas demonstraram capacidade para hidrolisar a ligação

éster. No entanto, quando comparadas com a LysB do fago Ms6 apresentaram sempre

uma menor atividade residual, embora os seus níveis dependessem do tamanho da cadeia

hidrocarbonada do substrato utilizado. Em linha com os resultados de modulação

estrutural, de entre as proteínas testadas, a LysB do fago Adjutor revelou os valores de

atividade residual mais baixo. Por último, a previsão efetuada a partir do modelo

estrutural da LysB do fago Ms6 do envolvimento dos resíduos Ser168, Asp249 e His318

na tríade catalítica foi confirmado por mutagénese dos resíduos alvo. A substituição por

Ala dos resíduos de Ser168 (S168A) e His318 (H318A) teve um efeito fortemente

negativo na atividade esterásica residual das variantes S168A (0.73±0.66%), H318A

(0.43±0.47%). As alterações conformacionais introduzidas na variante D249A deverão

ser tão drásticas que foi impossível recuperar a proteína na fração solúvel. Estes dados

reforçam o envolvimento da tríade Ser168-Asp249-His318 na catálise enzimática. Os

resultados experimentais apresentados na terceira parte do trabalho validam e dão força

ao modelo estrutural previsto para cada LysB, destacando e correlacionando as leves

variações de cada modelo relativamente à posição e orientação de cada resíduo catalítico

Ser, Asp e His.

Globalmente, os resultados apresentados nesta tese revelam a importância das

proteínas LysB na lise das micobactérias e constituem um importante contributo para a

conhecimento do modelo de lise mediada por micobacteriófagos.

Palavras-chave: micobacteriófago, bacteriófago, lise, Ms6, LysB, micobactéria,

envelope celular.

xv

ABSTRACT

Most bacteriophages, or simply phages, lyse bacterial hosts by producing at least

two proteins, holins and endolysins, which allow the release of the viron progeny at the

end of a lytic cycle. Holins are membrane proteins responsible for the lysis timing and

endolysins are enzymes that disrupt the peptidoglycan mesh. In addition to these two

essential lysis proteins, phages infecting Gram-negative hosts synthesize a third class of

lysis proteins, the spanins, necessary to overcome the bacterial outer membrane.

Mycobacteriophages, the phages that infect mycobacteria, also have to face an outer

membrane for phage release, although with a completely different composition from that

of Gram-negative bacteria. Most mycobacteriophages encode a protein, named LysB, an

enzyme with mycolyl-arabinogalactan esterase activity that hydrolyzes the link between

the outer membrane and the cell wall. In this work the mycobacteriophage Ms6 was used

as a model to investigate the role that LysB plays in lysis of Mycobacterium smegmatis.

It is shown that Ms6 LysB is an essential lysis protein for an efficient lysis of

mycobacteria. LysB proteins are highly diverse and some, including Ms6 LysB, have an

extended N-terminus. The presented results demonstrate the ability of this region to bind

to peptidoglycan, with a higher affinity for the M. smegmatis peptidoglycan, and absence

of this region affect the lysis progression and results in a faster rise period.

Based on comparative modeling and site directed mutagenesis the catalytic triad

involved in the esterase activity showed the catalytic triad formed by Ser168, Asp249 and

His318 to be essential for the esterase activity. A comparison with LysBs from other

mycobacteriophages shows that Ms6 LysB has the highest catalytic activity.

Overall, the results presented in this thesis define the importance of LysB in phage-

mediated lysis and further supports the notion that a complex host cell envelope can only

be overcome by an appropriate phage lytic toolkit.

Keywords: mycobacteriophage, bacteriophage lysis, Ms6, LysB, mycobacteria,

cell envelope

xvii

ABREVIATIONS

AG Arabinogalactan

AIDS Acquired Immunodeficiency Syndrome

Ala Alanine

Amp Ampicilin

AraLAM Lipoarabinomannan with no cap on the terminal Araƒ

Asp Aspartic acid

ATP Adenosine-5'-triphosphate

BCG Bacille Calmette-Guérin

BLAST Basic Local Alignment Search Tool

bp Base pair

BRED Bacteriophage recombineering of electroporated DNA

BSA Bovine serum albumin

CD Catalytic domain

CD4+ Cluster of differentiation 4

CM Cytoplasmic membrane

Cryo-EM Cryo-electron microscopy

CW Cell wall

CWBD Cell wall binding domain

DAP Diaminopimelic acid

DNA Deoxyribonucleic acid

ds Double stranded

DT Diphtheria toxin

Fig Figure

GC-MS Gas chromatography-mass spectrometry

GlcNac N-acetil glucosamine

GPLs Glycopeptidolipids

His Histidine

HIV Human immunodeficiency virus

ICTV International Committee on Taxonomy of Viruses

IPTG Isopropyl β-D-thiogalactoside

xviii

Kan Kanamycin

LAM Lipoarabinomanan

LB Luria-Bertani broth

LOS Lipooligosaccharide

LPS Lipopolysaccharide

mAGP Mycolil-arabinogalactan-peptidoglycan complex

ManLAM Mannose-capped lipoarabinomannan

ManLAM Lipoarabinomannan with mannose cap

MM Mycomembrane

NAG N-acetylglucosamine

NAM N-acetylmuramic acid

NCBI National Center for Biotechnology Information

OD Optical Density

OL Outermost layer

OM Outer membrane

PDIMs Phthiocerol dimycocerosates

PFU Plaque forming unit

PG Peptidoglycan

PGLs Phenolic glycolipids

PILAM Lipoarabinomannan with inositol phosphate cap

PIMs Phosphatidylinositol mannosides

pNP p-nitrophenyl

pNPB p-nitrophenyl butyrate

pNPL p-nitrophenyl laurate

pNPM p-nitrophenyl myristate

RNA Ribonucleic acid

SAR Signal-arrest-release

SDS Sodium dodecyl sulphate

Ser Serine

SNARE Soluble NSF attachment receptor

SP Signal peptide

ss Single stranded

TA Teichoic acid

xix

TB Tuberculosis

TCDB Transporter Classification Database

TDM Trehalose dimycolate

TLR Toll-like receptor

TMD Transmembrane domain

TMM Trehalose monomycolate

WHO World Health Organization

CHAPTER 1.

GENERAL INTRODUCTION

1. General Introduction

3

1.1 The bacteriophage

1.1.1 Phage discovery and ecology

In the beginning of the XX century, bacteriophages were discovered independently by

the British pathologist Frederick Twort in 1915 (Twort, 1915), and by the French–

Canadian microbiologist Félix d’Hérelle in 1917 (D’Herelle, 2007). Twort described the

“ultra-microscopic virus” of transparent micrococci colonies that could not grow (Twort,

1915), whereas d’Hérelle isolated an “anti-microbe” of Shigella and invented the term

“bacteriophage”, that derived from "bacteria" and the Greek: φαγεῖν (phagein) that means

"to devour" (D’Herelle, 2007). However, the first report of the lytic effect of a

bacteriophage was reported by the British bacteriologist Ernst Hankin in 1896 (Hankin,

1896). Although it is disputable who discovered the bacteriophages, the undeniable truth

is the impact that bacteriophages had on the development of molecular biology, and also

their relevance in biosphere at a planetary level. Every 48 hours, phages destroy about

half the bacteria in the world, a dynamic process that occurs in all ecosystems (Abedon,

2009).

Bacteriophages, the so-called phages, are viruses that infect bacteria and are estimated

to be 1030–1032 in the Earth biosphere, and 1023 is the expected number of phage

infections that occur each second (Hatfull et al., 2006, 2011). The bacteriophage

population overall diversity appears to be great: no genomically defined phage has been

isolated more than once, and the sequenced phage genomes are highly varied and

characterized by a high degree of mosaicism that likely arises from extensive horizontal

genetic exchange (Hendrix et al., 1999; Hatfull, 2008). The phage population is

exceptionally dynamic, turning over rapidly through constant selection and subsequent

amplification in permissive hosts (Pedulla et al., 2003). In each infection, the phage

encounters DNA - of bacterial or prophage origin - with which it can potentially

recombine to generate new genomic arrangements (Canchaya et al., 2003; Hatfull et al.,

2006; Kenzaka et al., 2010).

The potential of phages as a biotechnology tool (Marinelli et al., 2008) or therapeutic

agent has been the main issue of several studies that consider the usage of phages for


4

agriculture (Lam et al., 2018), aquaculture (Gon Choudhury et al., 2017), or even

bacterial infectious diseases in humans (Rohde et al., 2018) such as tuberculosis (Carrigy

et al., 2017).

1.1.2 Phage Classification and Characterization

The large diversity of morphological types of bacteriophages is used for taxonomy,

and it is based on their shape and size as well as on the type of nucleic acid (Figure 1.1.).

The present phage classification is derived from a scheme proposed by Bradley in 1967

(Bradley, 1997). In this classification six basic morphological types are included,

exemplified by phages T4, λ, T7, φX174, MS2, and fd. Virions can have a binary, cubic,

or helical symmetry or be pleomorphic. A few types have a lipid-containing envelope or

contain lipids as part of the viral particle (Ackermann, 2009). The phage capsid is a shell

of proteins (often in the shape of an icosahedron) that hold and protect the viral genome.

The nucleic acid can be a double-stranded (ds) DNA (as for the vast majority of phages),

dsRNA or single-stranded (ss) DNA or ssRNA. Phages may also have a tail, a helical

structure connected to the capsid, that can be short or long and this can be a contractile or

non-contractile structure. Tails may have additional structures, like baseplates, fibers and

spikes, which have a role in recognition and attachment to specific receptors present at

the surface of the host cell, usually connected to tail fibers containing at their tips

attachment sites that interact with receptors on the bacterial cell surface

The tailed phages constitute the order Caudovirales, which is divided into 3

families: Myoviridae, Siphoviridae, and Podoviridae, based on the tail

morphology. Myoviridae phages (T4, ϕKZ) are characterized by long straight contractile

tails, Siphoviridae phages (λ) possess long flexible non-contractile tails,

and Podoviridae phages (T7) have short non-contractile tails (Fokine et al., 2014). The

majority of discovered phages belong to these three families (Siphoviridae, Myoviridae

and Podoviridae), comprising 17 genera while the remaining phages are dispersed in 14

families, each with a small number of members (Ackermann, 2009).


5

Prokaryotic viruses classification grew over the years with the addition of new families

and genera. The International Committee on Taxonomy of Viruses (ICTV) at their last

report discriminate one order, includes six subfamilies, 80 genera, and 441 species

(Krupovic et al., 2016). Phage classification is open-ended since new phages are

discovered daily and the ICTV report updates (Adriaenssens et al., 2017).

Figure 1.1. Morphotypes of prokaryote viruses; Adapted from Ackermann et al. 2012.

1.1.3 Phage life cycles

Phages have diverse possible life cycles such as lytic, lysogenic, pseudolysogenic and

chronic infections (Weinbauer, 2004). Over the course of evolution, bacteriophages have


6

developed unique proteins that bind to and inactivate (or redirect), critical cellular

proteins in bacteria, shutting off key metabolic processes to divert the host metabolism to

the production of progeny phages (Liu et al., 2004).

The first step of phage infection is “Adsorption” to bacteria and it is shared by all

phages. The virus encounters its bacterial host during random motion and attaches to the

cell. For this, it is necessary that the phage recognizes bacterial receptors exposed on the

cell boundaries, that may be any of a wide variety of cell surface components, including

lipopolysaccharides, proteins, teichoic acids, peptidoglycan, flagella or pili (Guttman et

al., 2004; Bax, 2007; Rakhuba et al., 2010; Silva et al., 2016). Initially, the adsorption is

reversible but then becomes irreversible, which allows the phage to transfer the genetic

material into the host cell. The introduction of their genetic material into the cell is

accomplished, by the great majority of known phages, via ejection (Molineux et al.,

2013), however, for φ6-like phages, the ds-RNA is delivered by an endocytosis-like

mechanism (Romantschuk et al., 1988; Daugelavicius et al., 2005). The mechanisms of

ejection of the phage genome into the bacterial cell depends on the morphology of the

virus, but inevitably happens trough the tail (if present) and the hole in the bacterial cell

wall (Weinbauer, 2004; Bax, 2007). When the phage genome is inside de host cell,

depending on the phage life cycle, the infection progression takes different pathways.

Phages belonging to the Caudovirales order, which are about 96% of the reported so far,

may follow a lysogenic and/or a lytic cycle depending on whether they are temperate or

virulent respectively.

In the progression of the lytic pathway, the bacterial cell machinery is recruited to

replicate the viral genetic material and synthesize the viral proteins encoded by the foreign

nucleic acids, after which phage proteins generally self-assemble, packaging their genetic

material into capsids (Guttman et al., 2004). After enough virions have been produced,

and in order to release the virion progeny, the cell lyses, usually with the help of lytic

enzymes (Bernhardt et al., 2002; Catalão et al., 2013; Young, 2013) thereby killing the


7

Figure 1.2. Representation of the lysogenic and lytic cycle pathways of a temperate

phage. The bacterial cell (yellow) can be infected by the bacteriophage (purple) which

firsts adsorbs to the cell and then injects the viral DNA (orange). From this point the viral

DNA can follow a lysogenic or lytic pathway. The lytic cycle starts with the synthesis of

mRNA and early proteins. Next, the viral DNA starts to replicate and synthesis of late

proteins and viral components occurs. The following process is the assembly of viral

particles within the host cell. The final step of the lytic pathway is the lysis of the host cell

and release of viral progeny. The lysogenic cycle begins with the integration of the phage

DNA into the bacterial chromosome (black). After integration the host cell can divide

and propagates a stable copy of the bacterial and viral DNA to each daughter cell. This

propagation can cycle indefinitely or, due to specific circumstances, the viral DNA can

excise and enter the lytic cycle.


8

host cell. Phage mediated lysis is responsible for shaping bacterial population dynamics

and sometimes influencing their long term evolution, via a process designated as

generalized transduction (Weinbauer et al., 2004; Abedon, 2009). In addition to the lytic

cycle, temperate phages may undergo a lysogenic cycle, which, as opposed to the lytic

cycle, instead of killing their hosts, spend part of their life cycle in a quiescent state, either

by integrating into the host genome (like phage λ), or remaining like plasmids within the

host cytoplasm (like phage P1) (Salmond et al., 2015). During a lysogenic life cycle, the

genetic material of viral origin (prophage) may be maintained for thousands of

generations. Whenever the bacterial DNA replicates, the phage DNA will also replicate

at the same time, this allows that each daughter cell maintains a stable copy of the viral

genome. A repressor protein turns off the expression of other genes of the phage but

activates its own transcription. This prevents the expression of the lytic life cycle and

allows the lysogenic bacteria to divide and to transmit the prophage vertically (Gandon,

2016). The prophage may alter the bacteria phenotype by expressing genes that are not

expressed in the lytic infection progression in a process named lysogenic conversion. A

clinical relevant example of this phenomenon occurs in diphtheria infection, which is

caused by Corynebacterium diphtheriae. The disease symptoms are triggered by the DT

toxin and the gene that encodes DT (tox) is present in some corynephages, and

consequently, DT is only produced by C. diphtheriae isolates that harbor tox+ phages

(Holmes, 2000). Moreover, many bacteriophages convey virulence genes to the host cell,

shaping the bacteria by influencing bacterial adhesion, colonization, increasing bacterial

resistance to serum and phagocytosis, changing bacterial susceptibility to antibiotics or

encoding diverse bacterial exotoxins (Wagner et al., 2002; León et al., 2015). Regarding

important toxin genes acquired by transduction, besides cholera toxin, it is also of clinical

relevance the neurotoxin of Clostridium botulinum, the Shiga toxins found in Escherichia

coli O157 among others. (Wagner et al., 2002; León et al., 2015). Figure 1.2 represents

both lytic and lysogenic pathways of a temperate phage life cycle.

Besides the lytic and lysogenic cycles, phages may also have a psuedolysogenic

component to their life cycle. In a persistent infection (pseudolysogeny, phage carrier-

state) phages multiply only in a fraction of the population (Weinbauer, 2004).

Additionally, phages may also undergo a chronic infection. When a cell is infected and


9

phage progeny is constantly released from the host cell without lysing it, either by

budding or extrusion without obvious cell death, it is categorized as a chronic infection

(Weinbauer, 2004). The chronic infection is found in some archaeal viruses, in

filamentous phages, and in viruses that infect Mycoplasma (Clokie et al., 2011).

1.1.4 Mycobacteriophages, the viruses of mycobacteria

The interest in mycobacteriophages arises from the prevalent medical significance of

their hosts. M. tuberculosis the causal agent of tuberculosis (TB), has existed for millennia

and remains a major global health problem. This disease has the status of epidemy and in

2016 the world health organization (WHO), estimated that there were 1.4 million deaths

by TB, and an additional 0.4 million deaths resulting from TB disease among HIV-

positive people (WHO, 2017). These alarming numbers are enough to place TB on the

top 10 causes of death worldwide, ranking above HIV/AIDS as one of the leading causes

of death from an infectious disease (WHO, 2017). Mycobacteria are acid-fast staining

bacteria that can be readily divided in two groups based on their growth rate: slow-

growers such as M. tuberculosis and fast-growers such as M. smegmatis.

Mycobacteriophages were suggested to be useful for diagnosis of mycobacterial

infections, such as tuberculosis, in the development of tools for mycobacterial genetics

(Pedulla et al., 2003) and for analysis of mycobacterial peptidoglycan structure

(Mahapatra et al., 2013).

The collection of sequenced mycobacteriophage genomes offers insights into phages

that infect a common host, M. smegmatis mc2155 (Hatfull, 2012 a; b; Hatfull et al., 2013),

and because they share a common host, one can presume they are in genetic

communication with one another (Pedulla et al., 2003). At the time of this writing, more

than 9800 mycobacteriophages have already been isolated, most of them having M.

smegmatis as host, with over a total of 1600 genome sequences available in

http://www.phagesdb.org. Comparative genomics reveals mosaicism as the dominant

feature of phage genomes, suggesting that the phage population exchange segments

among its members (Hendrix et al., 1999; Pedulla et al., 2003). Nevertheless, simple

DNA comparison identifies groups of genomes more similar to one another than to other

phages, and these are referred to and grouped as clusters (Hatfull et al., 2006). So far,


10

mycobacteriophages have been assorted into a total of 26 different clusters, assorted from

A to Z which are further divided into sub clusters according to nucleotide sequence

similarities, and 6 singletons that have no close relatives (Hatfull, 2012 a; Hatfull et al.,

2013, 2016).

The complex cell envelope of M. tuberculosis encloses a wide and high content in

lipids with long chain fatty acids, which accounts for the highly hydrophobic cell surface

properties, resulting in increased resistance to dehydration and a natural protection against

environmental conditions and antibacterial drugs (Daffé et al., 1997). With this problem

in hands, it is imperative to study the core process by which the mycobacteriophages

accomplish the disruption of the mycobacteria cell envelope.

1.2 Cell barriers to phage release

To be released at the end of a lytic cycle, phages have to face the cell barriers, which

may be different or more or less complex according to the type of bacterial host.

There are different types of bacteria that have different cell envelopes, and every

simple variation, such as a substitution, covalent linkage, ramification or chain length of

a single component of the cell envelope can be a great challenge for the phage mediated

lysis. Therefore, it is important to disclosure briefly the major differences found between

the diverse layers of the cell envelope of Gram-negative, Gram-positive and mycobacteria

cells (Figure 1.3). The three main layers of the cell envelope are the cytoplasmic

membrane (CM) also known as inner membrane, which is surrounded by the cell wall

(CW), that in some bacteria is followed by an outer membrane (OM) and/or a capsule.

The idea of the host cell envelope as a barrier, in a phage mediated lysis perspective is

integrated in the comprehension of the phage lytic mechanisms.


11

Figure 1.3. Depiction of bacterial cell envelopes. (a) Gram-positive bacteria, (b) Gram-

negative bacteria; (c) mycobacteria. CM, cytoplasmic membrane; LA, lipoteichoic acids;

LAM, lipoarabinomannan; LP, lipoprotein; LPS, lipopolyssacharide; OM, outer

membrane; P, protein; PG, peptidoglycan; PIMs, phosphatidylinositol mannosides; PLs,

phospholipids; Po, porin; Pp, periplasm; TA, teichoic acids; TDM, trehalose dimycolate;

TMM, trehalose monomycolate. Figure from Catalão et al., 2013.


12

1.2.1 The cytoplasmic membrane

The CM is common to all bacteria and have the primordial role of confine the

cytoplasm and cellular components from escaping the cell. Bacterial survival is dependent

on the CM integrity, and is the lipid composition and homeostasis that allow the cell to

acclimatize to different environments (Zhang et al., 2008). The structure of bacterial

cytoplasmic membranes consist of proteins embedded in a lipid matrix made of

phospholipids, with two fatty acid chains (Zhang et al., 2008).

The difference in the concentration and charge of ions on each side of the CM

generates the proton motive force (pmf). The pmf is maintained by the cell and is required

to generate ATP, control bacterial autolysis and glucose transport. Any factor that disturbs

the pmf, depending on the extent, will impact the cell viability, and is said to depolarize

the cell (Novo et al., 1999). Is this form of energy that empowers all cellular reactions

and physiologic processes that allow cells to live and propagate (Weiner et al., 2007).

1.2.2 The cell wall - General considerations

The CW is a complex structure, present in almost all bacteria, that protects the bacteria

from the environment and surrounding milieu, but is also a tool to interact with the

environment as it enables the transport of information from the inside to the outside and

vice versa. Furthermore, the cell wall is a fine-tuned structure as it balances between

contradictions, providing the bacteria adequate impenetrability but allowing metabolism;

sufficient solidity but enough elasticity; strength to maintain shape and yet flexibility to

allow expansion and growth. It has important characteristics and is the subject of study

in diverse fields due to its tremendous impact in the cell biology. Although the CW

appears to be very complex, some common aspects are shared by many or most bacteria.

In 1884 Hans Christian Gram established the staining method that split in half the

bacterial world. The Gram-negative and Gram-positive bacteria differences are directly

related with the CW structure, the Gram-positive bacteria have a much thicker CW than

the Gram-negative bacteria. Besides, there are many other differences between the

components of the bacterial cell wall, specifically the peptidoglycan.


13

1.2.2.1 Peptidoglycan: the glycan strands and peptic cross-linking

The main component of the CW is the peptidoglycan (PG). The PG is long known as

an essential structural element, it is the result of the activity of more than 20 enzymes and

is well conserved between bacteria, especially those categorized as Gram-negative

(Schleifer et al., 1972).

The widespread presence of PG throughout the bacterial kingdom, and the fact that

animal cells have no homologous structure, makes it an excellent target for specific

antibacterial targeting, and it is indeed a common target in the mechanism of action of

many antibiotics. The PG assumes a vital role in the cell homeostasis, and inhibition of

PG biosynthesis or changes in its degradation during cell growth, will result in cell lysis

(Vollmer et al., 2008). Penicillin has been used for decades to target the enzymes involved

in the PG synthesis, and it has been efficient and used extensively, however due to many

different factors, increasing antibiotic resistance tends to turn this therapy inefficient has

a result of modifications in the biosynthetic enzymes.

The core structure of PG is made of glycan strands cross-linked by short peptides. The

glycan contains alternating chains of β-1,4 linked N-acetylglucosamine (NAG) and N-

acetylmuramic acid (NAM) units (Rogers et al., 1980). The NAM and NAG are cross-

linked by their peptide side chains. The glycan strands are the rigid fraction of the PG

polymer structure and are synthesized by transglycosylation reactions between

monomeric and disaccharide peptide units (Vollmer et al., 2008).

The length of the strands can vary greatly depending on the bacterial species and

growth conditions, although this is not correlated with the thickness of the PG (Vollmer

et al., 2008) Glycan strands average between 20 – 40 disaccharide units (Vollmer, 2008).

The majority of the variations occurs in the peptides of the PG and mostly in Gram-

positive bacteria (Schleifer et al., 1972). When considering only the glycan moiety of the

peptidoglycan it is remarkably uniform, however the glycan strands can be modified by

N-deacetylation, N-glycolylation and O-acetylation (Vollmer et al., 2008).


14

There are some important modifications of the glycan strands that can affect the

stability of the cell wall, increase the resistance to catalytic enzymes or enhance the

pathogenic profile (Vollmer, 2008). Many Bacillus species partially remove the acetyl

group of the NAG and NAM residues using N-deacetylases, while some other Gram-

positive bacteria, such as Micrococcus and Streptococcus, partially O-acetylate some

NAG residues, effectively inserting an extra acetyl group to create 2,6-N,O-diacetyl

muramic acid (Vollmer, 2008). Another modification, named N-glycolylation, changes

the acetyl group on the C2 of NAG to a glycolyl group. It was first described in M.

smegmatis (Petit et al., 1969) and is a characteristic of many Actinomycetales genera

(Raymond et al., 2005; Vollmer, 2008). The N-glycolylated PG seems to increase the

immunogenicity of M. tuberculosis, although it does not increase the pathogenicity

(Hansen et al., 2014). The existence and extension of all of these glycan strands

modifications depends on the growth phase and other environmental conditions. Slight

chemical modifications within the PG chain subtly change the properties of the cell wall,

for instances by increasing the stability or the resistance to lysozyme, a muramidase that

cleaves the PG between the NAM and NAG (Vollmer, 2008). Diverse studies show that

deacetylated PG is poorly targeted by the lysozyme, an enzyme that catalyzes the

hydrolysis of 1,4-beta-linkages between NAM and NAG in the peptidoglycan, and that

chemical acetylation of the PG restores the lysozyme activity on this substrate (Amano et

al., 1977; Westmacott et al., 1979; Vollmer et al., 2000).

The peptide cross-links are formed by transpeptidation reactions through two adjacent

peptide stems and are accountable for the flexibility of the PG sacculus (Vollmer et al.,

2008). The peptides and the way they cross-link may vary immensely which accounts for

more than 100 different peptidoglycans (Vollmer et al., 2008). The established system of

classification of peptidoglycans is based on structural data and is used since 1972

(Schleifer and Kandler, 1972). This classification takes in account the cross-linkage (A

or B), the presence or type of interpeptide bridge (1, 2 or 3) and the position of the peptide

stem (α, β or γ) in the peptidoglycan mesh (Vollmer, et al., 2008).

The usual, or more common, arrangement of amino acids (a.a.) in the peptide stem

linked to NAM is L-Ala‒D-Glu‒m-DAP‒D-Ala‒D-Ala, where m-DAP is meso-


15

diaminopimelic acid. This arrangement is found in nearly all Gram-negative bacteria, as

well as in the genera Bacillus and Mycobacterium (Schleifer et al., 1972). The Aγ1 group

can vary slightly and be divided in three types based on the position of the amidated

carboxyl group. E. coli and Bacillus megaterium have no amidation, while some of the

peptide subunits of Bacillus licheniformis are amidated at the α-carboxyl of D-Glu and of

B. subtilis are amidated on the free carboxyl group of m-DAP. The PG of Lactobacillus,

Corynebacterium diphtheriae, and Mycobacterium can be amidated at both locations

(Mahapatra et al., 2005).

As the PG structure may account for the differences between Gram-negative and

Gram-positive bacteria, there are more differences. The number, composition, and

structure of the components of the cell envelope diverge drastically between these

bacterial groups.

1.2.3 Gram-positive cell envelope

Bacteria belonging to the Gram-positive group have a thick layer of PG that constitutes

between 30% to 70% of the total CW weight and is the outermost barrier of the cell that

directly interacts with the environment. While it is not a true selective barrier, as is the

case with the OM or CM, the thick CW can prevent the incoming of larger molecules

(Shockman et al., 1983).

More recently, using transmission cryo-electron microscopy (cryo-EM), a bipartite

organization of the Gram-positive CW was observed, and it distinguishes an inner low

density zone and an outer high density zone (Matias et al., 2005, 2006; Zuber et al., 2008).

The inner low density zone has around 16-22 nm in Staphylococcus aureus and Bacillus

subtilis, and it looks similar to the periplasmic space of the Gram-negative bacteria

(Matias and Beveridge, 2005, 2006). It is suggested that many secreted proteins and wall

polymers may accumulate in this region (Vollmer, 2008). Also, several proteins that are

soluble in the periplasm of Gram-negative bacteria have lipid-modified homologues in

Gram-positive bacteria, for instances enzymes responsible for the PG turnover (Navarre

et al., 1999; Reith et al., 2011). The denser outer wall zone comprises the PG polymer


16

complex and ranges between 15 – 30 nm, as it may vary according to bacterial species,

growth phase, and environmental conditions (Vollmer et al., 2008).

Other non-PG cell wall polymers comprise 10-60% of the cell wall weight and mainly

includes polysaccharides such as teichoic acids (Schäffer et al., 2005). These anionic CW

polymers are covalently linked, via a phosphodiester bond, to the C6 of some NAM

(Navarre et al., 1999). Amongst the well-studied polysaccharides are: the teichoic acids

found in Bacillus, Staphylococcus, and Micrococcus (Shockman et al., 1983); the

choline-containing teichoic acids in Streptococcus pneumoniae (Hermoso et al., 2003; Di

Guilmi et al., 2017); and the arabinogalactan and lipoarabinomannan in mycobacteria

(Crick et al., 2001; Brennan, 2003; Yagi et al., 2003). These non-PG polysaccharides

play a role in cell growth, as many proteins that localize to the PG, such as PG-remodeling

enzymes like the Streptococcus pneumoniae Atl autolysin, bind only to the regions were

the PG is not hindered by the presence of wall teichoic acids (Giudicelli et al., 1984;

Schlag et al., 2010). Moreover, many of these polymers are associated to pathogenesis

and can cascade an immune response (Aderem et al., 2000).

1.2.4 Gram-negative cell envelope

Bacteria belonging to the Gram-negative group have only <10% of its total weight as

PG (Schleifer et al., 1972). On the outer face of the cytoplasmic membrane are the

periplasmatic space, the PG layer, and the asymmetrical OM (Matias et al., 2003). The

PG layer has anchored lipoproteins that spread along the OM. These lipoproteins are

covalently linked to the m-DAP C-terminal Lys or Arg residues, while the fatty acid

portion inserted into the inner leaflet of the OM (Höltje, 1998). The OM is a lipid bilayer

with an inner leaflet built of phospholipids. The outer leaflet has a different composition,

it contains membrane proteins like porins and notable lipopolysaccharides (LPS)

(Gronow et al., 2001). Approximately 75% of the cell surface is covered by the negatively

charged LPS, acting as a barrier against hydrophobic substances, while the rest of the

barrier halts the diffusion of hydrophilic substances (Gronow et al., 2001). The LPS have

three distinct parts: lipid A, the core and the O-antigen (Gronow et al., 2001; Raetz et al.,


17

2002). Regarding lipid A, it plays an important role in pathogenesis as an endotoxin

(Raetz et al., 2002). In humans, lipid A is recognized by the tool-like receptors (TLRs) of

the innate immune system, triggering an inflammatory response and recruitment of

phagocytic cells (Aderem et al., 2000).

1.2.5 Mycobacterial cell envelope

1.2.5.1 Global structure

The mycobacteria cell envelope is made of a CM, which has a structure common to

other bacteria, a CW with specific characteristics and an OM, despite the fact that

mycobacteria have been categorized as Gram-positive bacteria, due to the extensive

network of PG (Brennan et al., 1995).

The CW comprises a layer of PG covalently attached to arabinogalactan (AG), which

is in turn esterified at its nonreducing ends to α-alkyl, β-hydroxy long-chain (C60-C90) faty

acids, the mycolic acids. Altogether this is called the mycolil-arabinogalactan-

peptidoglycan (mAGP) complex, the CW core of mycobacteria. The mycolic acids, that

are covalently-linked to the mAGP, are part of the inner leaflet of a membrane bilayer,

the OM, where the outer leaflet is presumably composed by extractable lipids Hoffmann

et al., 2008; Zuber et al., 2008; Chiaradia et al., 2017). The exact arrangement of the

lipids in the outer membrane is still a matter of debate as a consequence of the methods

used for studying this structure.

With new microscopy techniques, it was possible to demonstrate the existence of this

OM, the so-called mycomembrane (MM) (Hoffmann et al., 2008; Zuber et al., 2008).

Interestingly, these cryo-EM observations found that the OM was not much thicker than

the CM; the mycobacterial OM was approximately 15% thicker than the CM (8 nm and

7 nm, respectively). Taking in consideration the thickness of this membrane and the very

long chain of mycolic acids, different models for the arrangement of the lipids in the OM

have been proposed. Zuber et al. (2008) have proposed that the mycolic acids chains fold

to stay within the OM inner leaflet. Hoffmann et al. (2008) have proposed two possible


18

models. The first one suggests that the mycolate chain of the mycolic acids extends into

the outer leaflet, and the second one that the mycolic acids remain in the periplasm with

only the ends extending into the inner leaflet.

The OM contains extractable (i.e. non-covalently bound to the cell) lipids and

lipoglycans: phosphatidylinositol mannosides (PIMs), phthiocerol dimycocerosates

(PDIMs), phenolic glycolipids (PGLs), a variety of acyltrehaloses, mannose-capped

lipoarabinomannan (ManLAM), trehalose monomycolates (TMM), trehalose

dimycolates (TDM), etc. (Jackson, M., 2014).

The organization and composition of the OM play a key role in the permeability of the

cell envelope and is responsible for the intrinsic resistance of mycobacteria to many

therapeutic agents and host defense mechanisms (Angala et al., 2014). Mycolic acids are

the hallmarks of mycobacteria, being present not only in all species of mycobacteria, but

also in related genera. Because mycolic acids are vital for the growth of mycobacterial

species the processes involving their biosynthesis have been one of the main targets for

antimicrobial drugs (Portevin et al., 2004). The OM is also supposed to contain porins for

the uptake of small hydrophilic molecules, as was shown for M. smegmatis with MspA

porins (Stahl et al., 2001).

Finally, the outermost layer (OL) present in mycobacteria surface is a loosely attached

capsular-like structure outside the MM. The OL of M. tuberculosis, was shown to have

polysaccharides and proteins with only minor amounts of lipids (2%–3%) (Sani et al.,

2010; Angala et al., 2014).

Figure 1.4 is a tentative model of the arrangement of the cell envelope of M. smegmatis

where all the above-mentioned layers are depicted.


19

Figure 1.4. Arrangement of M. smegmatis cell envelope. The plasma membrane (PM) is

around 7-8 nm protein and phospholipid bilayer membrane, separated from the cell wall

by a periplasmic space. The PG layer is covalently linked to the AG which in turn is

esterified by very long-chain fatty acids (mycolic acids) contained in the inner leaflet of

the MM. The outer leaflet of the MM is presumably composed of extractable lipids, which

include phospholipids, trehalose mycolates (TM), glycopeptidolipids, and lipoglycans.

The outermost layer (OL) is essentially constituted of proteins, a small amount of

carbohydrates and few lipids. Molecules are not drawn on scale. TMM: trehalose

monomycolates; TDM: trehalose dimycolates; GPL: glycopeptidolipids; PL:

phospholipids; PIM: phosphatidyl-myo-inositolmannosides; LAM: lipoarabinomannans;

TAG: triacylglycerols; Ag85: antigen 85. Figure from Chiaradia et al., 2017 with

permission.


20

1.2.5.2 Peptidoglycan modifications

Mycobacteria belong to the A1γ PG group, which is characterized by the presence of

m-DAP in the third position of the peptide stem (Schleifer et al., 1972). The m-DAP is

cross-linked to the D-Ala of another tetra-peptide moiety, but it is also possible the cross-

linking between two m-DAP moieties, and it was suggested that this provides additional

rigidity to the PG (Brennan et al., 1995). The cross-linking peptides are also more often

amidated on the Glu and m-DAP. Additionally, mycobacteria PG have N-glycolylated

muramic acids (MurNGly) instead of N-acetylated, a modification that provides more

opportunities for hydrogen interaction, thus adding stability to the PG (Brennan et al.,

1995; Raymond et al., 2005)

1.2.5.3 Arabinogalactan

Attached to the MurNGly residues are chains of arabinogalactan (AG). The AG is

composed of arabinose and galactose in the furanose form (Araf and Galf respectively)

(Brennan et al., 1995). With the work of Mcneil et al. (1991), using GC-MS analyses of

generated oligomers of AG it was possible to establish that: (i) the nonreducing termini

of arabinan consists of the structural motif [β-D-Araƒ- (1→2) –α-D-Araƒ]2–3,5-α-D-

Araƒ-(1→5)–α-D-(Araƒ6); (ii) the majority of the arabinan chains consist of 5-linked α-

D-Araƒ residues with branching introduced by 3,5-α-D-Araƒ; (iii) these chains are

attached to C-5 of some of the 6-linked Galƒ residues, and approximately two to three

arabinan chains are attached to the galactan core; (iv) the galactan regions consist of linear

alternating 5- and 6-linked β-D-Galƒ residues; (v) the galactan region of AG is linked to

the C-6 of some of the MurNGly residues of PG via a special diglycosyl-P bridge, α -L-

Rhaρ-(1→3)–D-GlcNAc- (1→P); (vi) and the mycolic acids are located in clusters of

four on the terminal hexaarabinofuranoside, but only two-thirds of these are mycolated.

The AG is very important in the mycobacteria cell envelope, in fact the majority of the

genes responsible for the biosynthesis of AG were shown to be essential in M.

tuberculosis and other mycobacteria (Kaur et al., 2009).


21

1.2.5.4 Lipoarabinomannan

The LAM is attached to the cell envelope and is mounted on a mannose chain

extending from the membrane phospholipid phospatidyl-myo-inositol (Briken et al.,

2004). The LAM exact localization is still a controversial matter, howerver it was

observed in both mycobacterial MM and CM fractions (Chiaradia et al., 2017).

The LAM is the mycobacteria equivalent of the lipoteichoic acids existent in the CW of

Gram-positive bacteria (Briken et al., 2004).

The arabinan chain of LAM is a chain and extends from the end of the mannose

backbone, and consists of α-(1→5) linked Araƒ with a tetra-arabinofuranoside and hexa-

arabinofuranoside branch (Chatterjee et al., 1991; Briken et al., 2004). The terminal Araƒ

has a variable capping depending on the mycobacterial species; M. smegmatis and other

fast-growing mycobacteria, terminate the arabinan with inositol phosphate caps (PILAM)

while M. tuberculosis and other pathogenic mycobacteria have one to three mannose as

caps (ManLAM); while Mycobacterium chelonae has been found to lack any capping

(AraLAM) (Briken et al., 2004). The differences in capping can influence pathogenesis;

ManLAM is shown to enhance the pathogeny of M. tuberculosis by inhibiting dendritic

cell function (Karakousis et al., 2004), and by inhibiting the activation of CD4+ human

T cells (Mahon et al., 2012).

1.2.5.5 Trehalose dimycolate and other extractable lipids

Numerous non-covalently linked or extractable lipids are found associated to the CW

of mycobacteria. According to Brennan et al.(1995), the most prevalent types of

extractable lipids include lipooligosaccharides (LOS), phenolic glycolipids,

glycopeptidolipids (GPLs), glycerophospholipids, sulfolipids and acetylated trehaloses,

such as the well-studied trehalose dimycolate (TDM), commonly known as cord factor.

As mentioned above, the extractable lipids are all located in the outer leaflet of the

OM, associated with the mycolic acid portion of the mAGP complex. As these lipids are

loosely packed they make a disordered low-density region (Brennan et al., 1995; Bansal-

Mutalik et al., 2014). The research done on these mycobacterial structural lipids indicates


22

that changes in their biosynthetic pathway can occur naturally and have impact in the

pathogenicity and infection progression of the bacilli (Minnikin et al., 2002; Karakousis

et al., 2004; Cambier et al., 2014). It is fair to assume that the primordial role of these

extractable lipids defines how the bacteria interacts with the surrounding environment

(Touchette et al., 2017).

The TDM is the most abundant glycolipid in the MM and, specifically for M.

tuberculosis, one of the major cell envelope components (Hunter et al., 2006). TDM is

formed by a disaccharide trehalose esterified to two mycolic acids, it has been the

subject of several studies and reviews that showed it is involved in inhibition of

phagosome maturation, macrophage death, cytokine production induction and

survivability of M. tuberculosis in macrophages (Rajni et al., 2011).


23

1.3 Overview of phage lysis players

As referred above, non-filamentous phages encode factors to compromise or destroy

the bacterial cell envelope at the end of the lytic cycle. Without an intact barrier, the

osmotic pressure causes the rupture of the host cell, so that its intracellular contents,

including the newly synthetized virions are released. Virulent phages can be categorized

accordingly to their genome size as small bacteriophages, with < 6 kb ssRNA or ssDNA

genomes, as opposed to large bacteriophages with dsDNA genomes.

Small bacteriophages encode a single protein to achieve the host cell lysis, presumably

because of the restricted coding capacity (Bernhardt et al., 2002). Only three different

prototypical lysis proteins have been found, and all of them from coliphages: E protein

from bacteriophage ΦX174 bacteriophage (ssDNA Microviridae); L protein from MS2

bacteriophage (ssRNA Leviviridae); and A2 protein from and Qβ bacteriophage (ssRNA

Alloleviviridae). Protein E functions as an inhibitor of phospho-MurNAc-pentapeptide

translocase (MraY) which is involved in the biosynthesis of bacterial murein (Bernhardt

et al., 2000, 2001; Zheng et al., 2009). Protein A2 blocks cell wall biosynthesis by

inhibiting UDP-N-acetylglucosamine enolpyruvyl transferase (MurA), an enzyme

catalyzing the first stage in PG synthesis (Brown et al., 1995), and for protein L the

mechanism of induced lysis in unknown (Reed et al., 2013). Although all these proteins

can achieve the lysis of the host cell they do it by a biosynthesis inhibitory mechanism,

and have no catalytic activity or degrading effect over the cell envelope.

Tailed phages constitute the clear majority of isolated bacterial viruses, meaning more

than 90% of all phages with completely sequenced genomes and available in NCBI

databases. To achieve lysis these phages synthetize at least one holin and an endolysin to

overcome the CM and the CW (Drulis-Kawa et al., 2015). Some phages may also encode

additional proteins to deal with the OM, such as spanins encoded in the genome of phages

infecting Gram-negative bacteria (Rajaure et al., 2015). Lysis genes are typically

clustered in the so-called “lytic cassette”. This is very important as lysis is an independent,

tightly regulated and temporarily scheduled pathway (Young, 2014), and deserve a

detailed description.


24

1.3.1 Holins: a molecular hourglass of lysis

Holins are essential proteins, in combination with endolysins, synthesized by dsDNA

bacteriophages to achieve host cell lysis. Holins are very diverse at the sequence level,

however they have common features: (i) when identified, usually their coding genes are

adjacent to the endolysin gene, (ii) are small hydrophobic proteins, (iii) have hydrophilic

and positively charged C-terminal, (iv) and have at least one transmembrane domain

(TDM) (Wang et al., 2000). The most recent database survey of Transporter

Classification Database (TCDB) have identified 52 holin families (Saier et al., 2015).

Holins are grouped into three main classes based on their number of TMD (Figure 1.5),

holins with 3 TMD belong to class I, with 2 TMD belong to class II and, finally, with

only one TMD belong to class I (Wang et al., 2000).

Figure 1.5. Schematic representation of the three classes of holins with 3, 2 or 1

transmembrane domain. Figure adapted from São-José et al., 2003.

Holin proteins are produced during the late stage of infection and form holes in the

cell membrane, but the purpose and consequence of those membrane lesions can vary.

Wang et al. (2000) have named the holins as “the protein clocks of bacteriophage

infections”, and this comparison means the holins are responsible for defining the timing

of host cell lysis. At that time, it was only thought that holins function was only to allow

the release of cytosol-accumulated endolysins through the pores formed. Nowadays it is

known that the endolysin does not always reach the CW through the CM holes formed by

the holins. Some holins form much smaller holes (“pinholes”) and are known as pinholins

(Park et al., 2007; Pang et al., 2009, 2013). These findings split the holin classification in

canonical holins and non-canonical holins (pinholins).


25

A canonical holin, such has λ phage S105, in the context of a phage infection, as it is

being synthesized, will progressively and harmlessly accumulate in the CM as mobile

oligomers, until it reaches a critical concentration (Young et al., 2014). At this point,

holins start to aggregate and form large rafts that permeabilize the CM and dissipate the

pmf, and it is proposed that the sudden alteration of charges cause conformational flipping

which allows the conversion of the holin aggregated rafts into micron-scale holes(Dewey

et al., 2010; To et al., 2014). These massive lesions allow the endolysin to escape from

the cytoplasm and attack the PG.

A pinholin pathway is somewhat similar to the one described for the canonical S105

holin. The protein is produced and accumulates gradually and harmlessly in the CM,

except that after triggering the rafts are smaller and in high numbers. Pang et al. (2009),

reported that the entire lambdoid phage 21 pinholin population of ≈7x103 molecules

oligomerize as heptamers, and forms ≈103 “pinholes” with an ≈2 nm estimated diameter

(Pang et al., 2009, 2013). These small holes are not large enough to allow the passage of

the endolysin, however they effectively depolarize the CM by dissipating the pmf. The

outcome leads inevitably to the host cell death but is essential to activate the endolysin

(Pang et al., 2009, 2013) (see below).

The concept of holins as clocks of phage infection remains true, however for both the

pinholins and canonical holins the comparison to an hourglass can be made. At a

biological defined time, the holin aggregates sets a timer for the lysis to occur, and like

the sand passes through the hourglass hole, so do the ions, flowing through the holin-

formed holes.

1.3.2 Antiholin: a post-translational lysis regulator

The antiholin function is to inhibit holin activity, acting as a holin antagonist in a post-

translational regulation mechanism. The antiholin function is well studied for the λ and

T4 phages.

The λ phage S gene encodes two membrane proteins, the S105 holin and the S107

antiholin (Bläsi et al., 1989). The S105 holin function depends on the three TMDs with a


26

N-out, C-in in topology. The S antiholin, S107, initiates at the first start codon of the S

gene (Met1) and has two additional residues at the N-terminus, Met and Lys, which are

absent in the S105. This causes the N-terminus of S107 to be positively charged and

consequently prevents the insertion of the first TMD into the CM, therefore S107 has no

holin function (White et al., 2010). The inhibitory effect of S107 arises from the

dimerization of S107 with S105, blocking the activation of the holin (Gründling et al.,

2000).

In T4, the holin T has only one TMD and a large periplasmic globular domain and the

antiholin-holin regulation is well studied. An antiholin RI inhibits holin activation by

binding to the holin periplasmic domain (Tran et al., 2005) and an additional protein, RIII

which is classified as a co-antiholin, binds to the cytoplasmic domain of the holin T

blocking the membrane hole formation (Chen et al., 2016). The system is quite peculiar

as it is the only described with a multi-antiholin regulation, and also it is capable of lysis

inhibition in response to environmental conditions such as the superinfection by other T4

virions (Doermann, 1948; Chen et al., 2016).

Although lambdoid phages have well studied holin-antiholin systems, two holin-like

genes have also been found on genomes of phages that infect Gram-positive bacteria two

holin-like genes (Catalão et al., 2013). Ms6, a phage that infects M. smegmatis, and

PBSX, a phage that infects B. subtilis, encode two holin-like proteins Gp4/Gp5 and

XhlA/XhlB, respectively (Krogh et al., 1998; Catalão et al., 2011 a). Inspired by the λ

and T4 lysis mechanism, some authors have speculated that these proteins might act as a

lambdoid holin/antiholin pair (Sheehan et al., 1999), however Krogh et al. (1998)

proposed that XhlB associate with the holin XhlA in the membrane to form the

“functional holin” that allows the endolysin (XlyA) release. The “functional holin”

concept is supported by the fact that pairwise combinations of XhlA or XhlB with the

XlyA do not trigger host cell lysis efficiently, and that only occurs when XhlA and XhlB

were co-expressed with the endolysin (Sheehan et al., 1999). Catalão et al. (2011a) have

also proposed that the pair of the Ms6 holin-like proteins Gp4/Gp5 act as a “functional

holin”, since the correct lysis timing was shown to be dependent on the interaction and

concerted action of those two membrane proteins (Catalão et al., 2011 a).


27

1.3.3 Endolysins: phage-encoded enzymes to degrade the peptidoglycan

As mentioned above, all dsDNA accomplish the host cell lysis do so by means of an

holin-endolysin coordinated action. Endolysins are phage-encoded enzymes designed to

digest specific bonds within the PG meshwork (Loessner, 2005), essential for the host

cell lysis process. The diversity of bacteria and bacteriophages suggest that we can expect

to find diverse endolysins, and indeed there are many, some of them were already studied

and predominant features were identified.

The endolysins are organized into five groups, according to the specific PG bond they

cleave (Figure 1.6.): N-acetyl-β-D-muramidases (lysozymes) and lytic transglycosylases

both cleaving the NAM-NAG bond, although the latter form a cyclic 1,6-anhydro-N-

acetylmuramic acid product; N-acetyl-β-D-glucosaminidases cutting the bond between

NAG and NAM; N-acetylmuramoyl-L-alanine amidases, which hydrolyze the amide

bond between NAM and L-alanine residues in the peptide chains; and endopeptidases,

which cleave the peptide bonds in the same chain (Loessner, 2005; Fischetti, 2010).

Endolysins may have a modular structure with one or more catalytic domain (CD), and

one or more cell wall binding domain (CWBD) (Loessner, 2005; Payne et al., 2012).

Typically, endolysins of phages that infect Gram-positive hosts and mycobacteria, have

a modular architecture with the CD located in the enzyme N-terminal region whereas the

CWBD is located closer to the C-terminal ( Loessner 2005). The majority of endolysins

from phages that infect Gram-negative have a globular structure with a single CD and no

distinct CWBD (Briers et al., 2007; Walmagh et al., 2012). It is known that some

endolysins added externally to Gram-positive bacteria can degrade the PG resulting in

immediate lysis (Fischetti, 2008). This feature encouraged the research on endolysins as

potential antibacterial agents, and this has been successfully demonstrated for some

endolysins (Fischetti, 2008, 2010, 2011; Diez-Martinez et al., 2015; Shen et al., 2016;

Thandar et al., 2016; Yang et al., 2017).


28

Figure 1.6. Representation of peptidoglycan bonds cleaved by the different endolysins

(indicated by blue arrows). Murein glycan strands consist of alternating NAG (N-acetyl-

D-glucosamine) and NAM (N-acetyl muramic acid) residues. Glycan strands are cross-

linked by short peptides. Figure adapted from Hanlon et al., 2007.


29

1.3.4 Spanins: proteins for membrane fusion

Counting holins, anti-holins and endolysins, spanins constitute a third functional class

of phage lysis proteins (Young, 2014). The spanins are the last player in the host lysis

pathway, exerting their function only after the holin disruption of the CM and endolysin

degradation of the PG (Berry et al., 2008, 2012).

The most extensively studied spanins are the Rz and Rz1 proteins of phage λ. The

extensive work of Ryland Young about the spanins dates back to 1979, the name Rz was

given due to the proximity to the R gene, but at that time only the Rz spanin was known

(Young et al., 1979). Rz1 was only found later, during an attempt to obtain purified Rz,

when Hanchy et al. (1993), observed the presence of an additional smaller protein

(Hanych et al., 1993). This phenomenon occurred because the cloned Rz fragment, due

to a +1 embedded second open reading frame, encodes both Rz and the unexpected

smaller protein that was named Rz1 (Hanych et al., 1993). Some years later, Kedzierska

et al. (1996), who was previously involved in Hanych work, studied specifically the Rz1

gene and demonstrated it encodes a 6.5-kDa pro-lipoprotein. This means the N-terminal

signal sequence is cleaved and undergoes a post-translational lipid modification to give

rise to a mature 40 a.a. OM lipoprotein (Kedzierska et al., 1996). At a molecular level Rz

and Rz1 are subunits of the spanin complex that spans the entire periplasmic space (Zhang

et al., 1999; Berry et al., 2008).

The concept of the importance that spanins play in λ lysis changed completely in 2012,

and are now considered to be essential for lysis (Berry et al., 2012). The demonstration

that the OM of Gram negative hosts is a significant barrier to the release of the progeny

virions, and that both the PG and the OM must be actively disrupted to achieve lysis, were

crucial to categorize the spanins as a third functional class of lysis proteins (Berry et al.,

2012; Young, 2014).

The current model suggested for the λ spanins action is topological, and it was shown

that the lysis occurs by a mechanism of CM–OM fusion (Rajaure et al., 2015). Rz has a

type-II topology (N-in, C-out) with a single N-terminal TMD that anchors in the CM, and

a periplasmic domain (Berry et al., 2008, 2010). The mature Rz1 lipoprotein anchors in


30

the inner leaflet of the OM (Kedzierska et al., 1996; Berry et al., 2008). These proteins

are subunits, designated as i-spanins (for inner) and o-spanins (for outer) respectively,

that interact with each other, via their periplasmic domains, to form a complex that spans

the entire periplasm, allowing the fusion of the inner and outer membranes, promoting

disruption of the latter (Berry et al., 2008).

Rz and Rz1 interaction occurs via their periplasmic domains after PG disruption (Berry

et al., 2008). The energy needed to fuse the CM and OM most likely derives from the

fusogenic properties of Rz/Rz1 as no free energy is available in the periplasm after holin

disruption of the pmf. Modifications in the tertiary and quaternary structure of the spanin

complex provides de energy necessary to fuse the CM and OM (Rajaure et al., 2015).

This mechanism is supported by further analysis of Rz/Rz1 that revealed subdomains that

resemble key motifs in established membrane-fusion systems: two coiled-coil domains in

Rz, a proline-rich region in Rz1 and flexible linkers in both proteins (Cahill et al., 2017

a). It was suggested that membranes are pulled into apposition by coiled-coils that

promote oligomerization and conformational change from extended to hairpin structure

(Cahill et al., 2017 a). This is known as “zippering model” that describes the progressive

assembly of an helical coiled structure proceeding toward membrane anchors, and was

designed to describe what happens in class I viral fusion and SNARE systems (Pobbati,

2006; Gao et al., 2012). A suppression analysis was conducted, with the goal of

identifying contacts between Rz and Rz1 and subdomains, and almost all suppressors

were clustered in a coiled-coil domain near the outer leaflet of the CM (Cahill et al., 2017

b). This reinforces the importance of the helical coiled structure to overcome the energy

barrier to fusion, further supporting the suggested fusion mechanism of Rz/Rz1 spanins

(Cahill et al., 2017 b).

The importance of spanins for lysis was strengthen by the identification of equivalent

spanin genes in most dsDNA phages of Gram-negative hosts (Summer et al., 2007). In

addition to λ , gene pairs like Rz/Rz1 were found in 43 other phages (Summer et al., 2007).

Although many of the gene pairs are organized as λ Rz/Rz1, in others, the second gene

extends beyond the end of the first gene or do not overlap at all. Additionally, there are

some phages where the spanin function is attributed to a single protein, having a N-


31

terminal OM lipoprotein determinant, and simultaneously anchoring to the CM by a C-

terminal transmembrane domain (TMD) (e.g.T1, gp11). These proteins were designated

u-spanins reflecting its unimolecular character (Young, 2014). Figure 1.7 shows the

current acknowledged model of spanins action.

Figure 1.7. Representation of the u-spanin and i-spanin/o-spanin complex within the

cell envelope. (A) Prior to lysis, the u-spanin and spanin complex accumulates trapped

within lacuna of the PG network. The endolysin progressively degrades the PG which

results in spanin activation. (B) The spanin complex undergoes a collapsing

conformational change, folding into a hairpin structure, a process known as zippering.

The juxtamembrane coiled-coil region becomes structured as the complex zippers into a


32

hairpin. The two-dimensional oligomerization of these complexes is not shown for figure

simplicity. The u-spanin is depicted establishing a beta-sheet from the predicted beta-

strand elements. Upon membrane fusion the phage release occurs. Figure adapted from

Young et al., 2014.

1.4 Phage-mediated host cell lysis

Many phages have a bacteriolytic behavior when the viral progeny is ready to be

released. Every aspect of the lysis of the host cell needs to be done in a precise and

coordinated mode. After disclosing the phage lysis players and the host barriers, this

section focuses on the mechanisms of phage-mediated cell lysis (Figure 1.8).

1.4.1 The λ phage as the lysis model paradigm

The very well-studied coliphage λ serves as a model for many dsDNA phages that use

a canonical holin-endolysin lysis mechanism.

The genes involved in λ lysis are clustered in a lysis cassette, transcribed from the pR’

late promoter, that contains four genes, S, R, Rz and Rz1, encoding five proteins

(Bernhardt et al., 2002). The dual-start motif of the gene S at codons 1 and 3, originate

S107 and S105, which are the antiholin and holin, respectively (Graschopf et al., 1999).

The R gene product codes for the endolysin, which is an 18-kDa soluble protein with

transglycosylase activity (Bieŉkowska-Szewczyk et al., 1981). The Rz and Rz1 genes

encode the i-spanin and the o-spanin, respectively, which interaction forms a complex

with the potential to span the periplasm (Berry et al., 2008; Cahill et al., 2017 a).

In the λ lysis model, the endolysin is synthesized and accumulates harmlessly in the

host cytoplasm in an active form, whereas the holin (S105) is accumulated in the CM

without disturbing its integrity. At a genetically defined time, the holin triggers and forms

large lesions that enable the R protein to diffuse into the periplasmic space, and attacks

the PG. These holes provide the pathway needed for endolysin release and are essential

for lysis (Young et al., 2000; Catalão et al., 2013). The timing of lysis is dependent on


33

the action of both the holin, S105 and of the anti-holin S107. As mentioned the Met-Lys

extension positive charge of S107 prevents the first TDM insertion in the CM. Further

reassuring the rapid integration of the S105 TDM1 in the CM, is the formylated Met

residue at the N-terminus of S105, whereas that of S107 is fully deformylated, as a

consequence of being exposed to the cytosol (White et al., 2010). The biologic ratio of

S107 and S105 produced is 1:2 and modifications on the balance of this ratio change the

timing of lysis, as increasing the proportion of S107 delay lysis (Bläsi et al., 1989).

However, after dissipation of membrane potential, the S107 TMD1 integrates in the CM,

thus functioning as an holin allowing the oligomerization and hole formation (White et

al., 2010). Between S107 and S105, the effect in vivo is determined by the existence of

the TDM1, since its deletion from S105 resulted in a lysis defect, with the mutant holin

not able to form holes, even in the presence of membrane poisons that dissipate the

membrane potential (White et al., 2010).

Although the holin trigger and the consequent dissipation of membrane potential is

enough to kill the host cell, the lysis only occurs due to the action of the endolysin R. To

allow the endolysin to escape from the cytosol to the periplasm, the hole formed by the S

holin is large enough to allow access of the endolysin to its PG substrate. Initial studies

demonstrated that the oligomerization of S105 was the trigger for membrane disruption

(Grundling et al., 2000). In vitro studies with purified protein S105 shown that S105

oligomers are ring-shaped structures that can further polymerize into head-to-tail

filaments (Savva et al., 2008). These ring structures eliminate the membrane potential

first, and it is suggested that holin rafts spontaneously convert to larger holes (Young,

2013). Previous studies demonstrated that the holins holes formed in vivo allow the

passage of an endolysin-β-galoctosidase fusion protein of 500 kDa (Wang et al., 2003).

Using cryo-EM it was determined that the average membrane holes diameter is 370nm

(Dewey et al., 2010).

In conclusion, the current knowledge allows to support a model in which the S105

trigger depends on the accumulation of S105 holin, forming oligomers with homo-typic

interactions in the CM, up to a moment when the S105 reaches a concentration that allows

membrane depolarization that drives S107 and S105 to oligomerize in hetero-typic


34

interaction, forming micron-scale interruptions in the CM (To et al., 2014). These lesions

are large enough to allow the endolysin R passage from the cytoplasm to the

peptidoglycan (Young, 2013).

After the holin and endolysin action, the last barrier to λ phage progeny release is the

OM. The Rz and Rz1 spanin units act at this step allowing the fusion of the CM and OM,

thus removing the last barrier that holds the virion progeny inside the host cell (Berry et

al., 2008, 2010). Without spanins function, lysis is blocked and progeny virions are

trapped in dead spherical cells, suggesting that the outer membrane has considerable

tensile strength (Young, 2014).

The current model for λ spanins action proposes that both Rz and Rz1 are synthetized

and accumulate as homodimers (Berry et al., 2013). Rz accumulates in the CM (i-spanin)

while Rz1 accumulates in the OM (o-spanin), however both spanin proteins are inhibited

by being trapped within the meshwork of the intact PG (Young, 2014). After PG

degradation by the R endolysin, the Rz highly structured periplasmic domain interacts

with the proline rich domain of Rz1, this causes the spanin complex to undergo a

conformational change, while forming helical structure via zippering that brings the

opposing membrane bilayers into proximity. It is this process that overcomes the energy

barrier and allows the fusion of the inner and outer membranes, and thus complete the

cell lysis (Berry et al., 2010; Rajaure et al., 2015; Cahill et al., 2017 a).


35

Figure 1.8. Endolysins activation and export models. (a) Endolysins wait in the

cytoplasm in an active form until collapse of the membrane potential and pore formation

by the holin, followed by passage of the endolysin to the PG. This is the mechanism of λ

phage. Endolysins holin-independent export is Sec-translocase dependent, and the

enzymes are exported gradually in an inactive form (b), accumulating in the periplasm,

as exemplified by endolysin of oenophage fOg44 or (c) remaining tethered to the CM by


36

a SAR sequence, like coliphage P1, waiting for the holin triggering and the membrane

potential collapse to get activated and degrade the PG. (d) In Ms6 the export of the full-

length endolysin (Lysin384) is independent of the holin, but dependent on the host sec

translocase and assisted by the chaperone Gp1. An N-terminally truncated version of the

Ms6 endolysin (Lysin241) is also synthesized, however the way this smaller endolysin is

exported to the extracytoplasmic environment is not known. Adapted from Catalão et al.,

2013.

1.4.2 Holin-independent export of endolysins

Some phages diverge from the classical, or canonical, λ model where the endolysin is

exported through the holes formed by the holin. If the endolysin export is not done

through the holes formed by the holin, then endolysins are exported independently

through an alternative mechanism.

1.4.3 fOg44, holin-independent export of Lys44 SP-endolysin

The first studies that revealed deviation from the λ paradigm in phage mediated host

cell lysis were performed with the temperate phage fOg44, which infects the Gram-

positive bacteria Oenococcus oeni (São-José et al., 2000). The fOg44 endolysin, Lys44

is a muramidase with a bona fide Sec-type signal peptide (SP) sequence, that engages the

general secretion pathway of bacteria (the Sec system) for its translocation to the CW,

with the subsequent removal of the SP, by a peptidase, to produce the active mature form

of the endolysin (São-José et al., 2000). Lys44 showed to be exported from the moment

of its synthesis, accumulating progressively at the site of action. It was suggested that

Lys44 accumulates in an inactive state in the PG layer, possibly due to the local conditions

that prevent the enzymatic action. Despite being exported in a holin-independent fashion,

phages encoding secreted endolysins also encode a holin and lysis does not occur until

holin trigger (São-José et al., 2000). fOg44 produces a functional holin (Hol44) whose

expression under λ native regulatory signals can complement a non-sense mutation in the

S holin gene (São-José et al., 2004). It was proposed that lysis occurs following


37

membrane disruption mediated by holin function. In fact, the use of agents that cause

nonselective permeabilization of the cell membrane, such as nisin and chloroform,

showed that, dissipation of the CM electrochemical gradient is necessary to sensitize O.

oeny and Lactococcus lactis to the endolysin Lys44, when it was added from the outside

or coming from within cells, respectively (Nascimento et al., 2008). This indicates that

even in the presence of a endolysin holin-independent export, the holin is still necessary

to determine the timing of lysis.

São-José et al. (2000) suggested that secretory lysins may be an evolutionary

advantage for phages infecting Gram-positive bacteria, since these have a much thicker

peptidoglycan. It seems reasonable to assume that an extensive lytic activity is required

to promote lysis, this way the progressive accumulation of endolysin in the PG would be

advantageous for a quick cell lysis (São-José et al., 2000).

1.4.4 SV1 holin-independent endolysin export

SV1 is a phage that infects the Gram-positive Streptococcus pneumoniae. The lysis

cassette is composed of two holin genes, svh1 and svh2, upstream of the lysin gene svl,

and a canonical holin-endolysin system was proposed for the SV1 induced lysis (López

et al., 2004). Recently, Frias et al. (2013) proposed that the endolysin (Svl) reaches the

PG through a novel holin-independent pathway. Indeed, in a holin-deficient background

the authors showed that export of the Sv1, which lack a recognizable secretory signal, is

dependent on the presence of choline, a structural component of the cell wall teichoic

acids and lipoteichoic acids. Since choline anchors Svl to the cell wall and TA precursors

may be loaded to choline intracellularly before being linked to the cell wall, it was

suggested that these could be involved in Svl translocation across the CM. However, the

exact mechanism of Svl export remains to be elucidated. Similarly to fOg44, once the

endolysin is positioned next to it target dissipation of the CM electrochemical gradient is

necessary and sufficient for endolysin activation leading to cell lysis, a role that is

assigned to the SV1 holin (Frias et al., 2013). Interestingly it was also shown that the SVl

holin forms holes large enough to allow the escape of Svl, thus, it is not clear if choline

is absolutely required for SV1 wt induced lysis.


38

Similarities between Svl and other pneumococcal phage endolysins, together with the

absence of signal sequences and ability to bind choline, led the authors to suggest that the

Svl particular translocation mechanism can be shared with the majority of pneumococcal

phages (Frias et al., 2013).

1.4.5 Holin-independent export of SAR-endolysins in coliphage systems

Similarly to what is described for fOg44, the coliphages P1 and 21 also have holin-

independent endolysin secretion (Xu et al., 2004). However, the endolysins LysP1 and

R21, encoded by phages P1 and 21, respectively, are considerably different from the

fOg44 endolysin Lys44 because they don’t have a recognizable cleavable SP sequence

(Xu et al., 2004). The N-terminus of LysP1 and R21 have a signal-arrest-release (SAR)

sequence, an unusual TMD that is characterized by the presence of weakly hydrophobic

amino acids flanked by charged residues (Xu et al., 2004). This SAR domain is not

cleaved like the SP sequence of Lys44, although it also allows the endolysin export

through the host Sec system, whereas in this case, the endolysin remains in the periplasm

tethered to the CM by the SAR sequence (Xu et al., 2004, 2005).

The proposed model of lysis with SAR endolysins also involves changes in the

membrane potential caused by the holins function. Briefly, the SAR endolysin is

synthesized and continuously secreted through the Sec system to the periplasm where it

accumulates tethered to the energized CM in its inactive conformation. At a genetically

defined time, the holin triggers, causing the depolarization of the CM which results in the

release of the endolysin from the membrane, which refolds into its catalytically active

form, resulting in lysis of the affected host (Xu et al., 2004).

The molecular mechanism of LysP1 activation and refolding was thoroughly studied,

and it involves changes in the sulfide bond formed between cysteine residue Cys44 and

Cys51, being the later located in the endolysin active site. When the SAR domain

disengage from the CM, following membrane depolarization by the holin function, it

exposes the Cys13 residue, the thiol group can engage an intramolecular thiol-di-sulfide

isomerization, creating a new sulfide bond connecting Cys13 and Cys44, allowing the

Cys51, and consequently the enzyme active site, to be accessible (Xu et al., 2004).


39

The activation mechanism of R21 does not involve any cysteine residue, in fact, the

SAR domain of R21 does not have a single cysteine. When the SAR domain releases from

the depolarized CM, the endolysin R21 refolds acquiring a different topology where the

SAR occupies a central position in the enzyme body, and also relocates one glutamate

residue that is essential to complete the Glu-Asp/Cys-Thr essential catalytic triad (Sun et

al., 2009). This mechanism enhances the importance of the SAR domain not only for the

secretion, and CM tethering keeping the enzyme in its inactive state, but also enrolls the

SAR domain in the activation of the enzyme by topological alteration and even the direct

participation in the catalytic process itself (Sun et al., 2009).

1.4.6 φKMV holin-independent SAR-endolysin export

The Pseudomonas aeruginosa phage φKMV lytic cassette encodes a typical pinholin

(KMV44) and an endolysin with characteristics of a SAR-endolysin (KMV45) (Briers et

al., 2011). The endolysin KMV45 has a non-cleaved Sec-dependent signal that is

essential for the catalytic activity of the endolysin, similarly to what is observed for phage

21 endolysin (Briers et al., 2011). Similarly to what happens for R21, KMV45 does not

have cysteine residue on the SAR-domain (Briers et al., 2011). Briers et al. (2011) suggest

that the initial blocking of the enzymatic activity of the RMV45 is likely to be due to

steric hindrance, then, the release from the membrane, may lead to a restructuring or

completion of the correct topology of the catalytic triad.

The importance of phage φKMV holin-endolisin model relays on the fact that it is the

archetype representative of “φKMV-like viruses” (Ceyssens et al., 2006; Kulakov et al.,

2009; Lammens et al., 2009). At least five other P. aeruginosa phages (LKD16, LUZ19,

φKF77, PT2 and PT5) share the characteristics of φKMV lytic cassette, consequently a

common signal-arrest-release lysis mechanism can be expected for these “φKMV-like

viruses” members (Briers et al., 2011).

1.4.7 ERA103, holin-independent export of Lys103 SAR-endolysin

Another example of a SAR endolysin export is that of phage ERA103 that infects the

Gram-negative bacterium Erwinia amylovora. ERA103 encodes Lys103, a SAR endolysin


40

with a unique activation mechanism. Similarly to LysP1, the SAR domain released from

the depolarized CM allows one cysteine residue, Cys12, to attack the disulfide bond

formed between Cys42 and Cys45. The Cys42-Cys45 disulfide bond maintains the enzyme

in its inactive state by blocking the catalytic Glu43. Once the SAR domain is released,

Lys103 is activated by thiol-disulfide isomerization driven by Cys12, exposing the Glu43

catalytic residue which was previously caged (Kuty et al., 2010). The endolysin Lyz103

was the first enzyme found to be regulated by disulfide bond caging of its active site.

1.5 The mycobacteriophage Ms6

Ms6 is a temperate mycobacteriophage that infects M. smegmatis, isolated from M.

smegmatis strain HB5688 in 1989 (Portugal et al., 1989). Accordingly to its morphology

and the type of DNA, a dsDNA of ~54 kb with an estimated G+C content of 61,5%, is

classified as a Siphoviridae family member (Portugal et al., 1989; unpublished data).

Portugal et al. (1989) reported that the phage particles are composed of an icosahedral

head with 80 nm in diameter and a long non-contractile tail 210 nm in length. However,

a recent electronic microscopy observation of Ms6 shows a isometric head approximately

54 nm in diameter and a tail approximately 188 nm in length (Figure 1.9) (Gigante A.,

unpublished) which is a commonly found morphology among the characterized

Siphoviridae mycobacteriophages (Hatfull et al., 2008).

Although the DNA sequence of Ms6 is not completely annotated, there are some

regions that have been the subject of several studies. The first studies identified the

genetic elements involved in integration of Ms6 genome into the host chromosomal DNA

(Figure 1.10). The integration of phage Ms6 occurs by a site-specific recombination

event, mediated by a 373 amino acid protein encoded by the integrase gene (int), between

the attP site localized near the 5‘ end of int and an attB site present within the 3’ end of

the tRNAAla gene of M. smegmatis (Freitas-Vieira et al., 1998). Near the integrase gene,

and transcribed in the opposite direction, is gene pin which encodes a protein involved in

a superinfection exclusion resistance mechanism. It was suggested that Pin inhibits a step

between adsorption of the phage and DNA injection (Pimentel, 1999).


41

The lysis cassette of Ms6 was also identified and shown to be composed by five genes

(Garcia et al., 2002). Characterization and function of each gene products has been lately

the subject of several studies. Due to sequence similarities of the lysis cassette with

members of cluster F, specifically F1 subcluster, Ms6 was suggested to be part of this

cluster (Catalão et al., 2011 b).

Figure 1.9. Electron microscopy image of Ms6 phage adsorbed to M. smegmatis cell.

Measurements in nm of tail length and 3-axis on capsid diameter. Uranyl acetate was used

as negative staining agent.


42

1.1.2. Genetic organization of the lysis module

The genetic organization and transcriptional control of the Ms6 lysis transcription unit

was described in 2002 by Garcia et al.. Transcription of all five genes, designated gp1

through gp5, is driven by a strong promoter region (Plys), found at the beginning of the

lysis locus. Downstream of Plys lies a 214 pb leader sequence harboring an intrinsic

transcription termination signal. The authors suggest that the regulation of Ms6 lysis

genes transcription may comprise an antitermination mechanism, and in order to allow

transcription to proceed beyond these termination signal, an antiterminator factor should

be synthesized (Garcia et al., 2002). This was the first transcription antitermination

mechanism reported in mycobacteriophages (Garcia et al., 2002).

Figure 1.10. Ms6 lytic cassette. The genetic organization is drawn to scale and each gene

name is indicated. The arrow indicates the direction of transcription. The promoter Plys is

upstream of gp1, separated by the leader sequence (L). (⁋) transcription termination

signal. Adapted with permission from Pimentel, M., 2014.

gp1, the first gene of the lysis cassette, encodes a small protein of 77 a.a. with 8.3 kDa,

Gp1 that shares the properties of molecular chaperones, particularly type III secretion

system chaperones. Gp1 interacts with the first 60 aa of Ms6 endolysin (LysA) N-

terminus and is involved in the endolysin export to the extracytoplasmic environment

(Catalão et al., 2010, 2011 b). The biological importance of Gp1 in the export of the

endolysin was demonstrated with the increase in alkaline phosphatase activity, resultant

from the translocation of the fusion protein LysA-PhoA only in the presence of Gp1 in

M. smegmatis (Catalão et al., 2010). A mutant Ms6 lacking gp1 gene was able to form

plaques, demonstrating that this gene is not essential for lysis, however a ~70% reduction

in the burst size was observed, meaning that Gp1 is required for an efficient lysis (Catalão

et al., 2010).


43

gp2 (or lysA) encodes a 1152 bp gene with a GTG start codon overlapping the gp1

TGA stop codon, in a different reading frame. It was shown that the endolysin has an N-

acetylmuramoyil-L-alanine amidase activity that cleaves the bond between the D-Mur and

L-Ala of mycobacterial PG (Mahapatra et al., 2013). Interestingly the Ms6 lysA gene

translates two fully functional endolysins: a full length lysin named Lysin384 reflecting

the number of amino acids, and Lysin241 which is a result of a second translation event

from a second GTG start codon in the same reading frame, positioned 426 bp downstream

of the first star codon (Catalão et al., 2011 c). Both peptides harbor the catalytic amidase

domain and both are necessary for a complete and efficient lysis (Catalão et al., 2011 c).

The biological importance of the endolysins Lysin384 and Lysin241 was evaluated by

Catalão et al. (2011 c). When M. smegmatis cells are infected with Ms6-Lysin384His6, a

phage mutant that encodes only Lysin384, lysis starts 90 min later, on the other hand, when

infected with a mutant Ms6-Lysin241His6 (a mutant that encodes only Lysin241), the lysis

starts 30 min later and less viral particles are released (Catalão et al., 2011 c). Noteworthy,

in absence of Gp1, the amount of Lysin384 detected following an infection with a Ms6

derivative mutant lacking gene gp1 dropped to residual levels (Catalão et al., 2011 b). It

was suggested that Gp1 could stabilize the endolysin through the interaction with the N-

terminus of Lysin384 (Catalão et al., 2010). The reason why Ms6 encodes two endolysins

and how is the Lysin241 exported, since it does not interact with the Gp1 (it lacks the first

143 aa), remains to be elucidated.

Lysin384 or Lysin241 were shown to be active from without, as E. coli crude extracts

containing these proteins were able to inhibit the growth of diverse bacteria when spotted

over inoculated agar medium (Catalão et al., 2011 c).

In addition to an endolysin, Ms6 also encodes a holin function. gp4 was shown to code

for a holin-like protein. Gp4 shares some structural characteristics with class II holins,

which are usually hydrophobic in nature and small in size, with a highly hydrophilic

carboxy-terminal domain and potential transmembrane domains (Catalão et al., 2011 a).

Although a holin function was supported also by its ability to complement a λ phage

S mutant, Gp4 does not function as a canonical holin. Co-expression of Ms6 LysA and


44

Gp4 in E. coli does not induce bacterial lysis (Garcia et al., 2002). However, when Ms6

gp4 was changed for the holin gene (gp11) of mycobacteriophage D29 a lysis phenotype

was obtained, suggesting that the holes formed by Gp4 are not large enough to allow the

passage of the Ms6 endolysin. This observation suggests that Gp4 may function as a

pinholin, also supported by the fact that, like the pinholin of phage 21, the first TMD of

Gp4 has characteristics of a SAR domain with a high percentage of weakly hydrophobic

and uncharged polar residues (Park et al., 2007; Pang et al., 2009; Catalão et al., 2011 a).

A deletion mutation of Ms6 lacking gp4 resulted in earlier lysis timing around 30 min, a

lysis phenotype that points towards antiholin function for Gp4 rather than holin(Catalão

et al., 2011 a).

gp5, the last gene of the lysis cassette encodes a 124 a.a. protein with a predicted TMD

at the N-terminus and a highly charged C-terminus, fitting the structural characteristics

of class III holins, however, it does not complement an S defective mutant λ phage.

Nevertheless, it was shown that Gp5 has a regulatory role in the timing of lysis, since a

deletion of gp5 from the Ms6 genome resulted in a viable phage, with a delayed time of

lysis (Catalão et al., 2011 a). Neither the putative class II holin nor the single TMD

polypeptide could trigger lysis in pairwise combinations with the endolysin LysA in E.

coli (Catalão et al., 2011 a). Additionally, cross-linking experiments showed that Ms6

Gp4 and Gp5 oligomerize and that both proteins interact. It was suggested that the correct

and programmed timing of lysis is achieved by the combined action of Gp4 and Gp5

(Catalão et al., 2011 a). A combined action for holin function is not unique to Ms6; a

similar action was proposed for the xhlA and xhlB gene products of prophage PBSX from

Bacillus subtilis (Krogh et al., 1998).

1.5.1 Ms6 LysB is a mycolyl-arabinogalactan esterase

The Ms6 lysB (gp3) encodes a protein of 332 a.a.. Analysis of the LysB deduced a.a.

sequence has revealed the presence of a conserved pentapeptide Gly-Tyr-Ser-Gln-Gly

motif, which matches the Gly-X-Ser-X-Gly consensus motif that is characteristic of

lipolytic enzymes (Jaeger et al., 1994, 1999; Bornscheuer, 2002). Gil, et al. (2008), have

experimentally demonstrated for the first time a lipolytic activity for a LysB protein and

showed that the Ms6 LysB is a mycolyl-arabinogalactan esterase cutting the bond


45

between mycolic acids of the outer membrane and the arabinogalactan polymer (Gil et

al., 2010).

The term ‘lipolytic enzymes’ comprises lipases (EC 3.1.1.3), carboxylesterases (EC

3.1.1.1) (Jaeger et al., 1999; Bornscheuer, 2002) and cutinases (EC 3.1.1.74) (Carvalho

et al., 1999; Longhi et al., 1999). Lipases are, by definition, enzymes that have the ability

to hydrolyze long-chain acylglycerols (≥C10), whereas esterases hydrolyze ester

substrates with short-chain fatty acids (≤C10) (Jaeger et al., 1999; Bornscheuer, 2002).

Cutinases hydrolyze the water-insoluble biopolyester cutin, a component of the waxy

exterior layer of plants. In addition to cutin, cutinase substrates include a wide variety of

esters ranging from soluble p-nitrophenyl (pNP) esters to insoluble long-chain

triglycerides.

Although the low overall sequence similarity of lipolytic enzymes and their

widespread molecular mass range (from 20 to 60 kDa), all these enzymes share a

comparable 3D fold, recognized as the α/β hydrolase fold (Holmquist, 2000). The activity

of these enzymes relies mainly on a catalytic triad formed by a nucleophilic Ser, an acidic

residue (Asp or Glu), and a His, in this order. Additionally, the Ser residue usually appears

in the conserved pentapeptide Gly-X-Ser-X-Gly consensus motif (Jaeger et al., 1999;

Longhi et al., 1999; Gupta et al., 2004).

Gil et al. (2008), determined the kinetic parameters of recombinant His6-LysB for the

hydrolysis of p-nitrophenyl esters, and the results showed that although this protein could

use a wide range of chain length substrates (C4–C18), it displayed a higher affinity for p-

nitrophenyl esters of longer chain length (C16 and C18). Moreover, using p-nitrophenyl

butyrate as a substrate, the recombinant LysB showed optimal activity at 23ºC and pH

7.5–8.0, and increased activity in the presence of 5mM Ca2+ or Mn2+ (Gil et al., 2008). In

addition to an mAGP esterase activity, Ms6 LysB can also target other mycolic acids-

containing lipids that are part of the mycobacterial cell envelope. It was shown that it also

hydrolyses trehalose dimycolate (TDM), a glycolipid involved in the virulence of

pathogenic species, and that this activity is not species specific since Ms6 LysB is able to

hydrolase the TDM of M. smegmatis, Mycobacterium bovis BCG and Mycobacterium

tuberculosis H37Ra (Gil et al., 2010). Due to the importance of the outer membrane as a


46

barrier to phage release at the end of a lytic cycle, it was proposed the main target of Ms6

LysB is the outer membrane linkage to the AG (Gil et al., 2010).

As has been described for phages that infect Gram-negative hosts, such as

bacteriophage λ, in order to overcome the OM of the host cell, these phages encode

spanins (Berry et al., 2008, 2012), and under the circumstance of a mycobacteriophage

infection, they too need to overtake the OM.

The importance of LysB in lysis is strengthen by the presence of LysB homologues in

most mycobacteriophages. In mycobacteriophage genomes the lysB is located close to

the lysA, while in phages ReqiPepy6 and ReqiDocB7 infecting the related genera

Rhodococcus equi they are not localized in the lysis cassette, and interestingly the

ReqiDocB7 phage lysB gene is transcribed in the opposite direction (Summer et al.,

2011). LysB proteins vary in length and a sequence alignment of mycobacteriophage

LysB proteins shows that they are highly diverse, with some having <20% identity,

(Payne et al., 2009). However they share an absolutely conserved residue, the active site

serine (position 168 in Ms6 LysB) (Payne et al., 2009). This suggests that these enzymes

are serine esterases, which typically have a Ser, Asp and His catalytic triad.

1.5.2 Ms6 current lysis model

It is possible to depict an Ms6 lysis model, based on the research performed on its lytic

operon and proteins, their interactions and the effect on the host cell. The model presented

here is not finished or definitive, and it is traced based on the current knowledge available

about Ms6, mycobacteriophages in general and other phages lysis mechanisms.

At the late stage of infection the two forms of LysA, both displaying peptidoglycan

hydrolase activity are synthesized.

As mentioned above, the lysA gene encodes two products LysA384 and the shorter

LysA241 and both display PG hydrolase activity. As LysA384 is synthesized, it is secreted

in a mechanism that involves the sec machinery of the host. The small protein Gp1

interacts with the endolysin N-terminus and assists the export (Catalão et al., 2010, 2011

b; c). The secreted endolysin remains inactive until drop of the pmf by the holins.


47

After being synthesized, the smaller LysA241 immediate fate is not yet known. If it

accumulates in the cytoplasm or is exported to the cell wall environment as long as it is

synthesized remains to be determined.

Meanwhile Gp4 and Gp5 proteins are also being produced and accumulate gradually

in the CM (Catalão et al., 2011 a). The combined action of Gp4 and Gp5 that triggers the

membrane potential collapse, defines the lysis timing and is predicted to activate the

endolysin function (Catalão et al., 2011 a).

After the disruption of the CM by the holin and the PG hydrolysis by the endolysin,

the mycobacterial OM is targeted by LysB. This mycolyl-arabinogalactan esterase can

detach the mycolic acids that are covalently linked to the AGP complex (Gil et al., 2010).

It is not known how LysB reaches its target.

The current Ms6 lysis model is likely a three-step process where the holins disorder

the CM, the endolysin hydrolyses the PG and then the LysB can disrupt the OM integrity,

a multi-step process that counterparts the complexity of the host cell envelope.


48

1.6 Thesis goals

This thesis title “Mycobacteriophage-mediated lysis: the role of LysB proteins” points

out the main goal of this work. The fact that LysB proteins, similarly to what happens

with spanins, constitute an additional class of proteins associated with phage-mediated

lysis, besides the classic and well documented holins and endolisins, focuses the attention

to them. LysB proteins participation and overall importance on the mycobacteria lysis is

not totally clear yet. It is important to mention that LysB proteins are known to have lipase

activity and act as a mycolyl-arabinogalactan esterase. In order to achieve the main goal,

specific objectives were outlined based on previous research about LysB proteins and

mycobacteriophage mediated lysis.

The research presented in this thesis is organized in three separated chapters, and each

have the objective to answer one key question:

1. What is the role of LysB on an M. smegmatis Ms6 infection cycle?

2. Is the N-terminus of Ms6 LysB a region with a peptidoglycan binding function?

3. In what way the Ms6 LysB esterase activity is comparable to LysB of Adjutor,

Trixie and U2 mycobacteriophages and which residues form the catalytic Ser-

Asp-His catalytic triad in Ms6 LysB?

The above-mentioned questions are of paramount importance for the understanding of

how mycobacteriophages achieve lysis of their hosts complex cell envelope.


49

1.7 References

Abedon, S.T., (2009) Phage evolution and ecology. Advances in applied microbiology.

Ackermann, H., (2009) Bacteriophages Methods and Protocols. Volume 1: Isolation,

Charterization, and Interactions. Methods in Molecular Biology.

Aderem, A., Ulevitch, R.J., (2000) Toll-like receptors in the induction of the innate

immune response. Nature., 406, 782–787.

Adriaenssens, E.M., Krupovic, M., Knezevic, P., Ackermann, H.W., Barylski, J., Brister,

J.R., Clokie, M.R.C., Duffy, S., Dutilh, B.E., Edwards, R.A., Enault, F., Jang, H.

Bin, Klumpp, J., Kropinski, A.M., Lavigne, R., Poranen, M.M., Prangishvili, D.,

Rumnieks, J., Sullivan, M.B. et al., (2017) Taxonomy of prokaryotic viruses: 2016

update from the ICTV bacterial and archaeal viruses subcommittee. Archives of

Virology., 162, 1153–1157.

Amano, K., Hayashi, H., Araki, Y., Ito, E., (1977) The action of lysozyme on

peptidoglycan with N-unsubstituted glucosamine residues. Isolation of glycan

fragments and their susceptibility to lysozyme. European journal of biochemistry.,

76, 299–307.

Angala, S.K., Belardinelli, J.M., Huc-Claustre, E., Wheat, W.H., Jackson, M., (2014) The

cell envelope glycoconjugates of Mycobacterium tuberculosis. Critical reviews in

biochemistry and molecular biology., 49, 361–399.

Bansal-Mutalik, R., Nikaido, H., (2014) Mycobacterial outer membrane is a lipid bilayer

and the inner membrane is unusually rich in diacyl phosphatidylinositol

dimannosides. Proceedings of the National Academy of Sciences., 111, 4958–4963.

Bax, R., (2007) Development of a twice daily dosing regimen of amoxicillin/clavulanate.

International Journal of Antimicrobial Agents., 30, 118–121.

Bernhardt, T.G., Roof, W.D., Young, R., (2000) Genetic evidence that the bacteriophage

ΦX174 lysis protein inhibits cell wall synthesis. Proceedings of the National

Academy of Sciences of the United States of America., 97, 4297–4302.

Bernhardt, T.G., Struck, D.K., Young, R., (2001) The lysis protein E of phi X174 is a

specific inhibitor of the MraY- catalyzed step in peptidoglycan synthesis. J Biol

Chem., 276, 6093–7.


50

Bernhardt, T.G., Wang, I.N., Struck, D.K., Young, R., (2002) Breaking free: “Protein

antibiotics” and phage lysis. Research in Microbiology., 153, 493–501.

Berry, J., Summer, E.J., Struck, D.K., Young, R., (2008) The final step in the phage

infection cycle: the Rz and Rz1 lysis proteins link the inner and outer membranes.

Molecular Microbiology., 70, 341–351.

Berry, J., Savva, C., Holzenburg, A., Young, R., (2010) The lambda spanin components

Rz and Rz1 undergo tertiary and quaternary rearrangements upon complex

formation. Protein Science., 19, 1967–1977.

Berry, J., Rajaure, M., Pang, T., Young, R., (2012) The spanin complex is essential for

lambda lysis. Journal of bacteriology., 194, 5667–5674.

Berry, J.D., Rajaure, M., Young, R., (2013) Spanin function requires subunit

homodimerization through intermolecular disulfide bonds. Molecular

microbiology., 88, 35–47.

Bieŉkowska-Szewczyk, K., Lipiŉska, B., Taylor, A., (1981) The R gene product of

bacteriophage λ is the murein transglycosylase. MGG Molecular & General

Genetics., 184, 111–114.

Bläsi, U., Nam, K., Hartz, D., Gold, L., Young, R., (1989) Dual translational initiation

sites control function of the lambda S gene. The EMBO journal., 8, 3501–3510.

Bornscheuer, U., (2002) Microbial carboxyl esterases: classification, properties and

application in biocatalysis. FEMS Microbiology Reviews., 26, 73–81.

Bradley, D.E., (1997) Ultrastructure of bacteriophages and bacteriocins. Bacteriol. Rev.,

230–314.

Brennan, P.., (2003) Structure, function, and biogenesis of the cell wall of Mycobacterium

tuberculosis. Tuberculosis., 83, 91–97.

Brennan, P.J., Nikaido, H., (1995) The envelope of mycobacteria. Annual review of

biochemistry., 64, 29–63.

Briers, Y., Volckaert, G., Cornelissen, A., Lagaert, S., Michiels, C.W., Hertveldt, K.,

Lavigne, R., (2007) Muralytic activity and modular structure of the endolysins of

Pseudomonas aeruginosa bacteriophages φKZ and EL. Molecular Microbiology.,

65, 1334–1344.


51

Briers, Y., Peeters, L.M., Volckaert, G., Lavigne, R., (2011) The lysis cassette of

bacteriophage фKMV encodes a signal-arrest-release endolysin and a pinholin.

Bacteriophage.

Briken, V., Porcelli, S.A., Besra, G.S., Kremer, L., (2004) Mycobacterial

lipoarabinomannan and related lipoglycans: From biogenesis to modulation of the

immune response. Molecular Microbiology.

Brown, E.D., Vivas, E.I., Walsh, C.T., Kolter, R., (1995) MurA (MurZ), the enzyme that

catalyzes the first committed step in peptidoglycan biosynthesis, is essential in

Escherichia coli. Journal of Bacteriology.

Cahill, J., Rajaure, M., O’Leary, C., Sloan, J., Marrufo, A., Holt, A., Kulkarni, A.,

Hernandez, O., Young, R., (2017)a Genetic Analysis of the Lambda Spanins Rz

and Rz1: Identification of Functional Domains. G3 (Bethesda, Md.)., 7, 741–753.

Cahill, J., Rajaure, M., Holt, A., Moreland, R., O’Leary, C., Kulkarni, A., Sloan, J.,

Young, R., (2017)b Suppressor Analysis of the Fusogenic Lambda Spanins.

Journal of Virology., 91, e00413-17.

Cambier, C.J., Takaki, K.K., Larson, R.P., Hernandez, R.E., Tobin, D.M., Urdahl, K.B.,

Cosma, C.L., Ramakrishnan, L., (2014) Mycobacteria manipulate macrophage

recruitment through coordinated use of membrane lipids. Nature., 505, 218–222.

Canchaya, C., Fournous, G., Chibani-Chennoufi, S., Dillmann, M.L., Brüssow, H., (2003)

Phage as agents of lateral gene transfer. Current Opinion in Microbiology.

Carrigy, N.B., Chang, R.Y., Leung, S.S.Y., Harrison, M., Petrova, Z., Pope, W.H.,

Hatfull, G.F., Britton, W.J., Chan, H.K., Sauvageau, D., Finlay, W.H., Vehring, R.,

(2017) Anti-Tuberculosis Bacteriophage D29 Delivery with a Vibrating Mesh

Nebulizer, Jet Nebulizer, and Soft Mist Inhaler. Pharmaceutical Research., 34,

2084–2096.

Carvalho, C.M.L., Aires-Barros, M.R., Cabral, J.M.S., (1999) Cutinase: From molecular

level to bioprocess development. Biotechnology and Bioengineering.

Catalão, M., Gil, F., Moniz-Pereira, J., Pimentel, M., (2011)a Functional analysis of the

Holin-Like proteins of mycobacteriophage Ms6. Journal of Bacteriology., 193,

2793–2803.


52

Catalão, M., Gil, F., Moniz-Pereira, J., Pimentel, M., (2011)b The endolysin-binding

domain encompasses the N-terminal region of the mycobacteriophage Ms6 Gp1

chaperone. Journal of Bacteriology., 193, 5002–5006.

Catalão, M.J., Gil, F., Moniz-Pereira, J., Pimentel, M., (2010) The mycobacteriophage

Ms6 encodes a chaperone-like protein involved in the endolysin delivery to the

peptidoglycan. Molecular Microbiology., 77, 672–686.

Catalão, M.J., Milho, C., Gil, F., Moniz-Pereira, J., Pimentel, M., (2011)c A second

endolysin gene is fully embedded in-frame with the lysA gene of

mycobacteriophage Ms6. (Abdelhay, E. S. F. W., Ed.)PLoS ONE., 6, e20515.

Catalão, M.J., Gil, F., Moniz-Pereira, J., São-José, C., Pimentel, M., (2013) Diversity in

bacterial lysis systems: Bacteriophages Show the way. FEMS Microbiology

Reviews., 37.

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. Journal of Bacteriology.

Chatterjee, D., Bozic, C.M., McNeil, M., Brennan, P.J., (1991) Structural features of the

arabinan component of the lipoarabinomannan of Mycobacterium tuberculosis. The

Journal of biological chemistry., 266, 9652–9660.

Chen, Y., Young, R., (2016) The Last r Locus Unveiled: T4 RIII Is a Cytoplasmic

Antiholin. (Christie, P. J., Ed.)Journal of Bacteriology., 198, 2448–2457.

Chiaradia, L., Lefebvre, C., Parra, J., Marcoux, J., Burlet-Schiltz, O., Etienne, G., Tropis,

M., Daffé, M., (2017) Dissecting the mycobacterial cell envelope and defining the

composition of the native mycomembrane. Scientific reports., 7, 12807.

Clokie, M.R.J., Millard, A.D., Letarov, A. V., Heaphy, S., (2011) Phages in nature.

Bacteriophage., 1, 31–45.

Crick, D.C., Mahapatra, S., Brennan, P.J., (2001) Biosynthesis of the arabinogalactan-

peptidoglycan complex of Mycobacterium tuberculosis. Glycobiology., 11, 107R–

118R.

D’Herelle, F., (2007) On an invisible microbe antagonistic toward dysenteric bacilli: brief

note by Mr. F. D’Herelle, presented by Mr. Roux. 1917. Research in microbiology.,

158, 553–554.


53

Daffé, M., Draper, P., (1997) The Envelope Layers of Mycobacteria with Reference to

their Pathogenicity. Advances in Microbial Physiology., Vol. 39.

Daffé, M., (2008) 1 The Global Architecture of the Mycobacterial Cell Envelope. The

Mycobacterial Cell Envelope., pp. 3–11.

Daugelavicius, R., Cvirkaite, V., Gaidelyte, A., Bakiene, E., Gabrenaite-Verkhovskaya,

R., Bamford, D.H., (2005) Penetration of enveloped double-stranded RNA

bacteriophages phi13 and phi6 into Pseudomonas syringae cells. J Virol., 79, 5017–

5026.

Dewey, J.S., Savva, C.G., White, R.L., Vitha, S., Holzenburg, A., Young, R., (2010)

Micron-scale holes terminate the phage infection cycle. Proceedings of the National

Academy of Sciences of the United States of America., 107, 2219–2223.

Di Guilmi, A.M., Bonnet, J., Peiβert, S., Durmort, C., Gallet, B., Vernet, T., Gisch, N.,

Wong, Y.S., (2017) Specific and spatial labeling of choline-containing teichoic

acids in Streptococcus pneumoniae by click chemistry. Chemical

Communications., 53, 10572–10575.

Diez-Martinez, R., De Paz, H.D., Garcia-Fernandez, E., Bustamante, N., Euler, C.W.,

Fischetti, V.A., Menendez, M., Garcia, P., (2015) A novel chimeric phage lysin

with high in vitro and in vivo bactericidal activity against Streptococcus

pneumoniae. Journal of Antimicrobial Chemotherapy., 70, 1763–1773.

Doermann, A.H., (1948) Lysis and Lysis Inhibition with Escherichia coli Bacteriophage.

Journal of bacteriology., 55, 257–276.

Drulis-Kawa, Z., Majkowska-Skrobek, G., Maciejewska, B., (2015) Bacteriophages and

Phage-Derived Proteins – Application Approaches. Current Medicinal Chemistry.,

22, 1757–1773.

Fischetti, V.A., (2008) Bacteriophage lysins as effective antibacterials. Current Opinion

in Microbiology., 11, 393–400.

Fischetti, V.A., (2010) Bacteriophage endolysins: A novel anti-infective to control Gram-

positive pathogens. International Journal of Medical Microbiology.

Fischetti, V.A., (2011) Exploiting what phage have evolved to control gram-positive

pathogens. Bacteriophage.

Fokine, A., Rossmann, M.G., (2014) Molecular architecture of tailed double-stranded

DNA phages. Bacteriophage., 4, e28281.


54

Freitas-Vieira, A., Anes, E., Moniz-Pereira, J., (1998) The site-specific recombination

locus of mycobacteriophage Ms6 determines DNA integration at the tRNA(Ala)

gene of Mycobacterium spp. Microbiology (Reading, England)., 144 ( Pt 1, 3397–

3406.

Frias, M.J., Melo-Cristino, J., Ramirez, M., (2013) Export of the pneumococcal phage

SV1 lysin requires choline-containing teichoic acids and is holin-independent.

Molecular microbiology., 87, 430–445.

Gandon, S., (2016) Why Be Temperate: Lessons from Bacteriophage λ. Trends in

Microbiology.

Gao, Y., Zorman, S., Gundersen, G., Xi, Z., Ma, L., Sirinakis, G., Rothman, J.E., Zhang,

Y., (2012) Single Reconstituted Neuronal SNARE Complexes Zipper in Three

Distinct Stages. Science., 337, 1340–1343.

Garcia, M., Pimentel, M., Moniz-Pereira, J., (2002) Expression of Mycobacteriophage

Ms6 Lysis Genes Is Driven by Two σ70-Like Promoters and Is Dependent on a

Transcription Termination Signal Present in the Leader RNA. Journal of

Bacteriology., 184, 3034–3043.

Gil, F., Catalão, M.J., Moniz-Pereira, J., Leandro, P., McNeil, M., Pimentel, M., Catalao,

M.J., Moniz-Pereira, J., Leandro, P., McNeil, M., Pimentel, M., (2008) The lytic

cassette of mycobacteriophage Ms6 encodes an enzyme with lipolytic activity.

Microbiology., 154, 1364–1371.

Gil, F., Grzegorzewicz, A.E., Catalão, M.J., Vital, J., McNeil, M.R., Pimentel, M., (2010)

Mycobacteriophage Ms6 LysB specifically targets the outer membrane of

Mycobacterium smegmatis. Microbiology (Reading, England)., 156, 1497–1504.

Giudicelli, S., Tomasz, A., (1984) Attachment of pneumococcal autolysin to wall teichoic

acids, an essential step in enzymatic wall degradation. Journal of Bacteriology.,

158, 1188–1190.

Gon Choudhury, T., Tharabenahalli Nagaraju, V., Gita, S., Paria, A., Parhi, J., (2017)

Advances in Bacteriophage Research for Bacterial Disease Control in Aquaculture.

Reviews in Fisheries Science and Aquaculture.

Graschopf, A., Bläsi, U., (1999) Functional assembly of the λ S holin requires periplasmic

localization of its N-terminus. Archives of Microbiology., 172, 31–39.


55

Gronow, S., Brade, H., (2001) Lipopolysaccharide biosynthesis: which steps do bacteria

need to survive? Journal of endotoxin research., 7, 3–23.

Grundling, A., Blasi, U., Young, R., (2000) Biochemical and Genetic Evidence for Three

Transmembrane Domains in the Class I Holin, S. Journal of Biological Chemistry.,

275, 769–776.

Gründling, A., Smith, D.L., Bläsi, U., Young, R., (2000) Dimerization between the holin

and holin inhibitor of phage λ. Journal of Bacteriology., 182, 6075–6081.

Gupta, R., Gupta, N., Rathi, P., (2004) Bacterial lipases: An overview of production,

purification and biochemical properties. Applied Microbiology and Biotechnology.

Guttman, B., Raya, R., Kutter, E., (2004) Basic Phage Biology. Bacteriophages. CRC

Press, pp. 29–63.

Hankin, E.H., (1896) L’action bactericide des eaux de la Jumna et du Gange sur le vibrion

du cholera. Annales de l’Institut Pasteur.

Hansen, J.M., Golchin, S.A., Veyrier, F.J., Domenech, P., Boneca, I.G., Azad, A.K.,

Rajaram, M.V.S., Schlesinger, L.S., Divangahi, M., Reed, M.B., Behr, M.A.,

(2014) N-glycolylated peptidoglycan contributes to the immunogenicity but not

pathogenicity of Mycobacterium tuberculosis. Journal of Infectious Diseases., 209,

1045–1054.

Hanych, B., Kȩdzierska, S., Walderich, B., Uznański, B., Taylor, A., (1993) Expression

of the Rz gene and the overlapping Rz1 reading frame present at the right end of

the bacteriophage lambda genome. Gene., 129, 1–8.

Hatfull, G.F., Pedulla, M.L., Jacobs-Sera, D., Cichon, P.M., Foley, A., Ford, M.E.,

Gonda, R.M., Houtz, J.M., Hryckowian, A.J., Kelchner, V.A., Namburi, S., Pajcini,

K. V., Popovich, M.G., Schleicher, D.T., Simanek, B.Z., Smith, A.L., Zdanowicz,

G.M., Kumar, V., Peebles, C.L. et al., (2006) Exploring the mycobacteriophage

metaproteome: Phage genomics as an educational platform. PLoS Genetics., 2,

0835–0847.

Hatfull, G.F., (2008) Bacteriophage genomics. Current Opinion in Microbiology.

Hatfull, G.F., Cresawn, S.G., Hendrix, R.W., (2008) Comparative genomics of the

mycobacteriophages: insights into bacteriophage evolution. Research in



56

Hatfull, G.F., Hendrix, R.W., (2011) Bacteriophages and their genomes. Current Opinion

in Virology., 1, 298–303.

Hatfull, G.F., (2012)a Complete Genome Sequences of 138 Mycobacteriophages. Journal

of Virology., 86, 2382–2384.

Hatfull, G.F., (2012)b The Secret Lives of Mycobacteriophages. Advances in Virus

Research., 82, 179–288.

Hatfull, G.F., Alliance, E., Hunters, P., Genomics, A., Sea-phages, E.S., Bailey, C.,

Barker, L.P., Bishai, W.R., Bow-, C., Bradley, K.W., Clarke, D., Eisenberg, N.,

Hendrix, R., Jacobs, W.R., Jacobs-sera, D., Jain, P., Larsen, M.H., Pope, W.H.,

Rubin, E., (2013) Complete Genome Sequences of 63 Mycobacteriophages.

Genome Announcements., 1, 1–3.

Hatfull, G.F., Science Education Alliance Phage Hunters Advancing Genomics and

Evolutionary Science (SEA-PHAGES) Program, KwaZulu-Natal Research

Institute for Tuberculosis and HIV (K-RITH) Mycobacterial Genetics Course,

University of California–Los Angeles Research Immersion Laboratory in Virology,

Phage Hunters Integrating Research and Education (PHIRE) Program, (2016)

Complete Genome Sequences of 61 Mycobacteriophages. Genome

Announcements., 4, e00389-16.

Hendrix, R.W., Smith, M.C., 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.

Hermoso, J.A., Monterroso, B., Albert, A., Galán, B., Ahrazem, O., García, P., Martínez-

Ripoll, M., García, J.L., Menéndez, M., (2003) Structural basis for selective

recognition of pneumococcal cell wall by modular endolysin from phage Cp-1.

Structure., 11, 1239–1249.

Hoffmann, C., Leis, A., Niederweis, M., Plitzko, J.M., Engelhardt, H., (2008) Disclosure

of the mycobacterial outer membrane: cryo-electron tomography and vitreous

sections reveal the lipid bilayer structure. Proceedings of the National Academy of

Sciences of the United States of America., 105, 3963–3967.

Holmes, R.K., (2000) Biology and Molecular Epidemiology of Diphtheria Toxin and the

tox Gene. The Journal of Infectious Diseases., 181, S156–S167.


57

Holmquist, M., (2000) Alpha Beta-Hydrolase Fold Enzymes Structures, Functions and

Mechanisms. Current Protein and Peptide Science., 1, 209–235.

Höltje, J. V, (1998) Growth of the stress-bearing and shape-maintaining murein sacculus

of Escherichia coli. Microbiology and molecular biology reviews : MMBR., 62,

181–203.

Hunter, R.L., Olsen, M.R., Jagannath, C., Actor, J.K., (2006) Multiple roles of cord factor

in the pathogenesis of primary, secondary, and cavitary tuberculosis, including a

revised description of the pathology of secondary disease. Annals of clinical and

laboratory science., 36, 371–386.

Jaeger, K.E., Ransac, S., Dijkstra, B.W., Colson, C., Heuvel, M., Misset, O., (1994)

Bacterial lipases. FEMS Microbiology Reviews., 15, 29–63.

Jaeger, K.E., Dijkstra, B.W., Reetz, M.T., (1999) Bacterial Biocatalysts: Molecular

Biology, Three-Dimensional Structures, and Biotechnological Applications of

Lipases. Annual Review of Microbiology., 53, 315–351.

Jackson M. (2014). The mycobacterial cell envelope-lipids. Cold Spring Harb. Perspect

Med. 4:a021105.

Karakousis, P.C., Bishai, W.R., Dorman, S.E., (2004) Mycobacterium tuberculosis cell

envelope lipids and the host immune response. Cellular microbiology., 6, 105–116.

Kaur, D., Guerin, M.E., Skovierová, H., Brennan, P.J., Jackson, M., (2009) Chapter 2:

Biogenesis of the cell wall and other glycoconjugates of Mycobacterium

tuberculosis. Advances in applied microbiology., 69, 23–78.

Kedzierska, S., Wawrzynów, A., Taylor, A., (1996) The Rz1 gene product of

bacteriophage lambda is a lipoprotein localized in the outer membrane of

Escherichia coli. Gene., 168, 1–8.

Kenzaka, T., Tani, K., Nasu, M., (2010) High-frequency phage-mediated gene transfer in

freshwater environments determined at single-cell level. The ISME journal., 4,

648–659.

Kothe, H.W., Vobis, G., Kroppenstedt, R.M., Henssen, A., (1989) A Taxonomic Study

of Mycolateless, Wall Chemotype IV Actinomycetes. Systematic and Applied


Krogh, S., Jørgensen, S.T., Devine, K.M., (1998) Lysis genes of the Bacillus subtilis

defective prophage PBSX. Journal of bacteriology., 180, 2110–2117.


58

Krupovic, M., Dutilh, B.E., Adriaenssens, E.M., Wittmann, J., Vogensen, F.K., Sullivan,

M.B., Rumnieks, J., Prangishvili, D., Lavigne, R., Kropinski, A.M., Klumpp, J.,

Gillis, A., Enault, F., Edwards, R.A., Duffy, S., Clokie, M.R.C., Barylski, J.,

Ackermann, H.W., Kuhn, J.H., (2016) Taxonomy of prokaryotic viruses: update

from the ICTV bacterial and archaeal viruses subcommittee. Archives of Virology.,

161, 1095–1099.

Kulakov, L.A., Ksenzenko, V.N., Shlyapnikov, M.G., Kochetkov, V. V., Del Casale, A.,

Allen, C.C.R., Larkin, M.J., Ceyssens, P.J., Lavigne, R., (2009) Genomes of

“phiKMV-like viruses” of Pseudomonas aeruginosa contain localized single-

strand interruptions. Virology.

Kuty, G.F., Xu, M., Struck, D.K., Summer, E.J., Young, R., (2010) Regulation of a phage

endolysin by disulfide caging. Journal of Bacteriology., 192, 5682–5687.

Lam, P., Steinmetz, N.F., (2018) Plant viral and bacteriophage delivery of nucleic acid

therapeutics. Wiley Interdisciplinary Reviews: Nanomedicine and

Nanobiotechnology., 10, e1487.

Lammens, E., Ceyssens, P.J., Voet, M., Hertveldt, K., Lavigne, R., Volckaert, G., (2009)

Representational Difference Analysis (RDA) of bacteriophage genomes. Journal of

Microbiological Methods.

León, M., Bastías, R., (2015) Virulence reduction in bacteriophage resistant bacteria.

Frontiers in Microbiology., 06.

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. Nature biotechnology., 22, 185–191.

Loessner, M.J., (2005) Bacteriophage endolysins - Current state of research and

applications. Current Opinion in Microbiology., 8, 480–487.

Longhi, S., Cambillau, C., (1999) Structure-activity of cutinase, a small lipolytic enzyme.

Biochimica et Biophysica Acta (BBA) - Molecular and Cell Biology of Lipids.,

1441, 185–196.

López, R., García, E., (2004) Recent trends on the molecular biology of pneumococcal

capsules, lytic enzymes, and bacteriophage. FEMS Microbiology Reviews., 28,

553–580.


59

Mahapatra, S., Yagi, T., Belisle, J.T., Espinosa, B.J., Hill, P.J., McNeil, M.R., Brennan,

P.J., Crick, D.C., (2005) Mycobacterial lipid II is composed of a complex mixture

of modified muramyl and peptide moieties linked to decaprenyl phosphate. Journal

of Bacteriology., 187, 2747–2757.

Mahapatra, S., Piechota, C., Gil, F., Ma, Y., Huang, H., Scherman, M.S.M.S., Jones, V.,

Pavelka, M.S.M.S.M.S., Moniz-Pereira, J., Pimentel, M., McNeil, M.R.M.R.,

Crick, D.C.D.C., (2013) Mycobacteriophage Ms6 LysA: a peptidoglycan amidase

and a useful analytical tool. Applied and environmental microbiology., 79, 768–

773.

Mahon, R.N., Sande, O.J., Rojas, R.E., Levine, A.D., Harding, C. V., Henry Boom, W.,

(2012) Mycobacterium tuberculosis ManLAM inhibits T-cell-receptor signaling by

interference with ZAP-70, Lck and LAT phosphorylation. Cellular Immunology.,

275, 98–105.

Marinelli, L.J., Piuri, M., Swigoňová, Z., Balachandran, A., Oldfield, L.M., van Kessel,

J.C., Hatfull, G.F., (2008) BRED: A Simple and Powerful Tool for Constructing

Mutant and Recombinant Bacteriophage Genomes. (Baker, S. E., Ed.)PLoS ONE.,

3, e3957.

Matias, V.R.F., Al-Amoudi, A., Dubochet, J., Beveridge, T.J., (2003) Cryo-transmission

electron microscopy of frozen-hydrated sections of Escherichia coli and

Pseudomonas aeruginosa. Journal of Bacteriology., 185, 6112–6118.

Matias, V.R.F., Beveridge, T.J., (2005) Cryo-electron microscopy reveals native

polymeric cell wall structure in Bacillus subtilis 168 and the existence of a

periplasmic space. Molecular Microbiology., 56, 240–251.

Matias, V.R.F., Beveridge, T.J., (2006) Native cell wall organization shown by cryo-

electron microscopy confirms the existence of a periplasmic space in

Staphylococcus aureus. Journal of Bacteriology., 188, 1011–1021.

McNeil, M., Daffe, M., Brennan, P.J., (1991) Location of the mycolyl ester substituents

in the cell walls of mycobacteria. Journal of Biological Chemistry., 266, 13217–

13223.

Minnikin, D.E., Kremer, L., Dover, L.G., Besra, G.S., (2002) The Methyl-Branched

Fortifications of Mycobacterium tuberculosis. Chemistry & Biology., 9, 545–553.


60

Molineux, I.J., Panja, D., (2013) Popping the cork: mechanisms of phage genome

ejection. Nature reviews. Microbiology., 11, 194–204.

Nascimento, J.G., Guerreiro-Pereira, M.C., Costa, S.F., São-José, C., Santos, M.A.,

(2008) Nisin-triggered activity of Lys44, the secreted endolysin from Oenococcus

oeni phage fOg44. Journal of Bacteriology., 190, 457–461.

Navarre, W.W., Ton-That, H., Faull, K.F., Schneewind, O., (1999) Multiple enzymatic

activities of the murein hydrolase from staphylococcal phage phi11: Identification

of a D-alanyl-glycine endopeptidase activity. Journal of Biological Chemistry.,

274, 15847–15856.

Novo, D., Perlmutter, N.G., Hunt, R.H., Shapiro, H.M., (1999) Accurate flow cytometric

membrane potential measurement in bacteria using diethyloxacarbocyanine and a

ratiometric technique. Cytometry., 35, 55–63.

Pang, T., Savva, C.G., Fleming, K.G., Struck, D.K., Young, R., (2009) Structure of the

lethal phage pinhole. Proceedings of the National Academy of Sciences of the

United States of America., 106, 18966–18971.

Pang, T., Fleming, T.C., Pogliano, K., Young, R., (2013) Visualization of pinholin lesions

in vivo. Proceedings of the National Academy of Sciences of the United States of

America., 110, E2054-63.

Park, T., Struck, D.K., Dankenbring, C.A., Young, R., (2007) The pinholin of lambdoid

phage 21: Control of lysis by membrane depolarization. Journal of Bacteriology.,

189, 9135–9139.

Payne, K., Sun, Q., Sacchettini, J., Hatfull, G.F., (2009) Mycobacteriophage Lysin B is a

novel mycolylarabinogalactan esterase. Molecular Microbiology., 73, 367–381.

Payne, K.M., Hatfull, G.F., (2012) Mycobacteriophage endolysins: Diverse and modular

enzymes with multiple catalytic activities. PLoS ONE., 7.

Pedulla, M.L., Ford, M.E., Houtz, J.M., Karthikeyan, T., Wadsworth, C., Lewis, J.A.,

Jacobs-Sera, D., Falbo, J., Gross, J., Pannunzio, N.R., Brucker, W., Kumar, V.,

Kandasamy, J., Keenan, L., Bardarov, S., Kriakov, J., Lawrence, J.G., Jacobs,

W.R., Hendrix, R.W. et al., (2003) Origins of Highly Mosaic Mycobacteriophage

Genomes. Cell., 113, 171–182.

Petit, J.F., Adam, A., Wietzerbin-Falszpan, J., Lederer, E., Ghuysen, J.M., (1969)

Chemical structure of the cell wall of Mycobacterium smegmatis. I — Isolation and


61

partial characterization of the peptidoglycan. Biochemical and Biophysical

Research Communications., 35, 478–485.

Pimentel, M., (1999) Biologia molecular da resistência à superinfecção por

bacteriófagos: estudo do gene pin do micobacteriófago Ms6. Univ. of Lisbon.

Pimentel, M., (2014) Genetics of Phage Lysis. Microbiology Spectrum., 2, 1–13.

Pobbati, A. V., (2006) N- to C-Terminal SNARE Complex Assembly Promotes Rapid

Membrane Fusion. Science., 313, 673–676.

Portevin, D., De Sousa-D’Auria, C., Houssin, C., Grimaldi, C., Chami, M., Daffé, M.,

Guilhot, C., (2004) A polyketide synthase catalyzes the last condensation step of

mycolic acid biosynthesis in mycobacteria and related organisms. Proceedings of

the National Academy of Sciences of the United States of America., 101, 314–319.

Portugal, I., Anes, E., Moniz-Pereira, J., (1989) Temperate Mycobacteriophage from M.

smegmatis. Acta Leprogica., 1, 243–244.

Raetz, C.R., Whitfield, C., (2002) Lipopolysaccharide endotoxins. Annual review of


Rajaure, M., Berry, J., Kongari, R., Cahill, J., Young, R., (2015) Membrane fusion during

phage lysis. Proceedings of the National Academy of Sciences of the United States

of America., 112, 5497–5502.

Rajni, Rao, N., Meena, L.S., (2011) Biosynthesis and Virulent Behavior of Lipids

Produced by Mycobacterium tuberculosis : LAM and Cord Factor: An Overview.

Biotechnology Research International.

Rakhuba, D. V., Kolomiets, E.I., Szwajcer Dey, E., Novik, G.I., (2010) Bacteriophage

receptors, mechanisms of phage adsorption and penetration into host cell. Polish

Journal of Microbiology.

Raymond, J.B., Mahapatra, S., Crick, D.C., Pavelka, M.S., (2005) Identification of the

namH gene, encoding the hydroxylase responsible for the N-glycolylation of the

mycobacterial peptidoglycan. The Journal of Biological Chemistry., 280, 326–333.

Reed, C.A., Langlais, C., Wang, I.N., Young, R., (2013) A2 expression and assembly

regulates lysis in Qβ infections. Microbiology (United Kingdom)., 159, 507–514.

Reith, J., Mayer, C., (2011) Peptidoglycan turnover and recycling in Gram-positive

bacteria. Applied Microbiology and Biotechnology., 92, 1–11.


62

Rogers, H.J., Perkins, H.R., Ward, J.B., (1980) Microbial Cell Walls and Membranes.

Springer Netherlands, Dordrecht.

Rohde, C., Resch, G., Pirnay, J.P., Blasdel, B.G., Debarbieux, L., Gelman, D., Górski,

A., Hazan, R., Huys, I., Kakabadze, E., Łobocka, M., Maestri, A., Almeida,

G.M.D.F., Makalatia, K., Malik, D.J., Mašlaňová, I., Merabishvili, M., Pantucek,

R., Rose, T. et al., (2018) Expert Opinion on Three Phage Therapy Related Topics:

Bacterial Phage Resistance, Phage Training and Prophages in Bacterial Production

Strains. Viruses., 10.

Romantschuk, M., Olkkonen, V.M., Bamford, D.H., (1988) The nucleocapsid of

bacteriophage phi 6 penetrates the host cytoplasmic membrane. The EMBO

journal., 7, 1821–1829.

Saier, M.H., Reddy, B.L., (2015) Holins in bacteria, eukaryotes, and archaea:

Multifunctional xenologues with potential biotechnological and biomedical

applications. Journal of Bacteriology.

Salmond, G.P.C., Fineran, P.C., (2015) A century of the phage: past, present and future.

Nature reviews. Microbiology., 13, 777–786.

Sani, M., Houben, E.N.G., Geurtsen, J., Pierson, J., de Punder, K., van Zon, M., Wever,

B., Piersma, S.R., Jiménez, C.R., Daffé, M., Appelmelk, B.J., Bitter, W., van der

Wel, N., Peters, P.J., Van Wel, N. Der, Peters, P.J., (2010) Direct visualization by

Cryo-EM of the mycobacterial capsular layer: A labile structure containing ESX-

1-secreted proteins. (Ramakrishnan, L., Ed.)PLoS Pathogens., 6, e1000794.

São-José, C., Parreira, R., Vieira, G., Santos, M.A., (2000) The N-terminal region of the

Oenococcus oeni bacteriophage fOg44 lysin behaves as a bona fide signal peptide

in Escherichia coli and as a cis-inhibitory element, preventing lytic activity on

oenococcal cells. Journal of bacteriology., 182, 5823–5831.

São-José, C., Santos, S., Nascimento, J., Brito-Madurro, A.G., Parreira, R., Santos, M.A.,

(2004) Diversity in the lysis-integration region of oenophage genomes and evidence

for multiple tRNA loci, as targets for prophage integration in Oenococcus oeni.

Virology., 325, 82–95.

Savva, C.G., Dewey, J.S., Deaton, J., White, R.L., Struck, D.K., Holzenburg, A., Young,

R., (2008) The holin of bacteriophage lambda forms rings with large diameter.



63

Schäffer, C., Messner, P., (2005) The structure of secondary cell wall polymers: How

Gram-positive bacteria stick their cell walls together. Microbiology.

Schlag, M., Biswas, R., Krismer, B., Kohler, T., Zoll, S., Yu, W., Schwarz, H., Peschel,

A., Götz, F., (2010) Role of staphylococcal wall teichoic acid in targeting the major

autolysin Atl. Molecular Microbiology., 75, 864–873.

Schleifer, K.H., Kandler, O., (1972) Peptidoglycan types of bacterial cell walls and their

taxonomic implications. Bacteriological Reviews., 36, 407–477.

Sheehan, M.M., Stanley, E., Fitzgerald, G.F., Van Sinderen, D., (1999) Identification and

characterization of a lysis module present in a large proportion of bacteriophages

infecting Streptococcus thermophilus. Applied and Environmental Microbiology.,

65, 569–577.

Shen, Y., Barros, M., Vennemann, T., Gallagher, D.T., Yin, Y., Linden, S.B., Heselpoth,

R.D., Spencer, D.J., Donovan, D.M., Moult, J., Fischetti, V.A., Heinrich, F.,

Lösche, M., Nelson, D.C., (2016) A bacteriophage endolysin that eliminates

intracellular streptococci. eLife., 5.

Shockman, G.D., Barrett, J.F., (1983) Structure, function, and assembly of cell walls of

gram-positive bacteria. Annu Rev Microbiol., 37, 501–527.

Silva, J.B., Storms, Z., Sauvageau, D., (2016) Host receptors for bacteriophage

adsorption. FEMS Microbiology Letters.

Stahl, C., Kubetzko, S., Kaps, I., Seeber, S., Engelhardt, H., Niederweis, M., (2001)

MspA provides the main hydrophilic pathway through the cell wall of

Mycobacterium smegmatis. Molecular Microbiology., 40, 451–464.

Summer, E.J., Berry, J., Tran, T.A.T., Niu, L., Struck, D.K., Young, R., (2007) Rz/Rz1

Lysis Gene Equivalents in Phages of Gram-negative Hosts. Journal of Molecular

Biology., 373, 1098–1112.

Summer, E.J., Liu, M., Gill, J.J., Grant, M., Chan-Cortes, T.N., Ferguson, L., Janes, C.,

Lange, K., Bertoli, M., Moore, C., Orchard, R.C., Cohen, N.D., Young, R., (2011)

Genomic and functional analyses of Rhodococcus equi phages ReqiPepy6,

ReqiPoco6, ReqiPine5, and ReqiDocB7. Applied and Environmental



64

Sun, Q., Kuty, G.F., Arockiasamy, A., Xu, M., Young, R., Sacchettini, J.C., (2009)

Regulation of a muralytic enzyme by dynamic membrane topology. Nature

structural & molecular biology., 16, 1192–1194.

Thandar, M., Lood, R., Winer, B.Y., Deutsch, D.R., Euler, C.W., Fischetti, V.A., (2016)

Novel Engineered Peptides of a Phage Lysin as Effective Antimicrobials against

Multidrug-Resistant Acinetobacter baumannii. Antimicrobial Agents and

Chemotherapy., 60, 2671–2679.

To, K.H., Young, R., (2014) Probing the structure of the S105 hole. Journal of


Touchette, M.H., Seeliger, J.C., (2017) Transport of outer membrane lipids in

mycobacteria. Biochimica et Biophysica Acta - Molecular and Cell Biology of

Lipids., 1862, 1340–1354.

Tran, T.A.T., Struck, D.K., Young, R., (2005) Periplasmic Domains Define Holin-

Antiholin Interactions in T4 Lysis Inhibition. Journal of Bacteriology., 187, 6631–

6640.

Twort, F.W., (1915) An Investigation on the Nature of Ultra-Microscopic Viruses. The

Lancet., 186, 1241–1243.

Vollmer, W., Tomasz, A., (2000) The pgdA gene encodes for a peptidoglycan N-

acetylglucosamine deacetylase in Streptococcus pneumoniae. Journal of Biological

Chemistry., 275, 20496–20501.

Vollmer, W., Blanot, D., De Pedro, M.A., (2008) Peptidoglycan structure and

architecture. FEMS Microbiology Reviews., 32, 149–167.

Vollmer, W., (2008) Structural variation in the glycan strands of bacterial peptidoglycan.

FEMS Microbiology Reviews., 32, 287–306.

Wagner, P.L., Waldor, M.K., (2002) Bacteriophage control of bacterial virulence.

Infection and Immunity.

Walmagh, M., Briers, Y., Dos Santos, S.B., Azeredo, J., Lavigne, R., (2012)

Characterization of Modular Bacteriophage Endolysins from Myoviridae Phages

OBP, 201varphi2-1 and PVP-SE1. PloS one., 7, e36991.

Wang, I.N., Smith, D.L., Young, R., (2000) Holins: The Protein Clocks of Bacteriophage

Infections. Annual Review of Microbiology., 54, 799–825.


65

Wang, I.N., Deaton, J., Young, R., (2003) Sizing the holin lesion with an endolysin-β-

galactosidase fusion. Journal of Bacteriology., 185, 779–787.

Weinbauer, M.G., (2004) Ecology of prokaryotic viruses. FEMS Microbiology Reviews.

Weinbauer, M.G., Rassoulzadegan, F., (2004) Are viruses driving microbial

diversification and diversity? Environmental Microbiology.

Weiner, J.H., Rothery, R.A., Weiner, J.H., Rothery, R.A., (2007) Bacterial Cytoplasmic

Membrane. Encyclopedia of Life Sciences. John Wiley & Sons, Ltd, Chichester,

UK, pp. 1–10.

Westmacott, D., Perkins, H.R., (1979) Effects of lysozyme on Bacillus cereus 569:

rupture of chains of bacteria and enhancement of sensitivity to autolysins. Journal

of general microbiology., 115, 1–11.

White, R., Tran, T.A.T., Dankenbring, C.A., Deaton, J., Young, R., (2010) The N-

terminal transmembrane domain of λ S is required for holin but not antiholin

function. Journal of Bacteriology., 192, 725–733.

WHO, (2017) WHO Global tuberculosis report 2017. World Health Organization Press.

Xu, M., Struck, D.K., Deaton, J., Wang, I.N., Young, R., (2004) A signal-arrest-release

sequence mediates export and control of the phage P1 endolysin. Proceedings of

the National Academy of Sciences., 101, 6415–6420.

Xu, M., Arulandu, A., Struck, D.K., Swanson, S., Sacchettini, J.C., Young, R., (2005)

Disulfide Isomerization After Membrane Release of Its SAR Domain Activates P1

Lysozyme. Science., 307, 113–117.

Yagi, T., Mahapatra, S., Mikušová, K., Crick, D.C., Brennan, P.J., (2003) Polymerization

of mycobacterial arabinogalactan and ligation to peptidoglycan. Journal of

Biological Chemistry., 278, 26497–26504.

Yang, H., Zhang, H., Wang, J., Yu, J., Wei, H., (2017) A novel chimeric lysin with robust

antibacterial activity against planktonic and biofilm methicillin-resistant

Staphylococcus aureus. Scientific reports., 7, 40182.

Young, R., Way, J., Way, S., Yin, J., Syvanen, M., (1979) Transposition mutagenesis of

bacteriophage lambda. A new gene affecting cell lysis. Journal of Molecular

Biology., 132, 307–322.

Young, R., Wang, I.N., Roof, W.D., (2000) Phages will out: Strategies of host cell lysis.

Trends in Microbiology.


66

Young, R., (2013) Phage lysis: do we have the hole story yet? Current opinion in


Young, R., (2014) Phage lysis: Three steps, three choices, one outcome. Journal of


Zhang, N., Young, R., (1999) Complementation and characterization of the nested Rz and

Rz1 reading frames in the genome of bacteriophage λ. Molecular and General

Genetics., 262, 659–667.

Zhang, Y.M., Rock, C.O., (2008) Membrane lipid homeostasis in bacteria. Nature

Reviews Microbiology., 6, 222–233.

Zheng, Y., Struck, D.K., Young, R., (2009) Purification and functional characterization

of φX174 lysis protein E. Biochemistry., 48, 4999–5006.

Zuber, B., Chami, M., Houssin, C., Dubochet, J., Griffiths, G., Daffe, M., (2008) Direct

Visualization of the Outer Membrane of Native Mycobacteria and Corynebacteria.

Journal of Bacteriology., 190, 5672–5680.

CHAPTER 2.

THE MS6 MYCOLYL-ARABINOGALACTAN

ESTERASE LYSB IS ESSENTIAL

FOR AN EFFICIENT

MYCOBACTERIOPHAGE-INDUCED LYSIS

This chapter contains data published in:

Gigante, A. M., Hampton, C. M., Dillard, R. S., Gil, F., Catalão, M. J., Moniz-Pereira, J.,

Wright, E. R., Pimentel, M., (2017) The Ms6 Mycolyl-Arabinogalactan Esterase LysB is

Essential for an Efficient Mycobacteriophage-Induced Lysis. Viruses., 9, 343.

doi:10.3390/v9110343

The Ms6 Mycolyl-Arabinogalactan Esterase LysB is Essential

for an Efficient Mycobacteriophage-Induced Lysis

Adriano M. Gigante 1, Cheri M. Hampton 2, Rebecca S. Dillard 2, Filipa Gil 1,

Maria João Catalão 1, José Moniz-Pereira 1, Elizabeth R. Wright 2 and Madalena

Pimentel 1

1 Research Institute for Medicines (iMed.ULisboa), Faculdade de Farmácia, Universidade de Lisboa,

Lisbon, 1649-003, Portugal;

2 Division of Pediatric Infectious Diseases, Emory University School of Medicine, Children’s

Healthcare of Atlanta, Atlanta, GA, 30345, USA;

Running Title: Ms6 LysB Function

Keywords: bacteriophage lysis; mycobacteriophage; Ms6; LysB; mycobacteria; spanins; cryo-electron

microscopy

Viruses

(Volume 9, Issue 11, 2017)

2. Ms6 LysB Function

71

Acknowledgments

We would like to thank Varalakshmi Vissa and Michael McNeil (Colorado State

University, Fort Collins, CO, USA) for providing pVVAP vector and the Robert P.

Apkarian Integrated Electron Microscopy Core, Emory University for microscopy

services and support. This work was supported in part by Fundação para a Ciência e

Tecnologia (FCT-MCES, Portugal) Grant PTDC/IMI-MIC/0694/2012 to MP; Emory

University, Children’s Healthcare of Atlanta, and the Georgia Research Alliance to

E.R.W.; the Center for AIDS Research at Emory University (P30 AI050409); public

health service grant GM104540 to E.R.W. from the NIH/NIGMS, and NSF grant

0923395 to E.R.W. AG (SFRH/BD/87685/2012) is a recipient PhD fellowship from FCT-

MCES, Portugal.

Conflicts of Interest:

The authors declare no conflict of interest. The founding sponsors had no role in the

design of the study; in the collection, analyses, or interpretation of data; in the writing of

the manuscript; or in the decision to publish the results.


73

Abstract

All dsDNA phages encode two proteins involved in host lysis, an endolysin and a holin

that target the peptidoglycan and cytoplasmic membrane, respectively. Bacteriophages

that infect Gram-negative bacteria encode additional proteins, the spanins, involved in

disruption of the outer membrane. Recently, a gene located in the lytic cassette was

identified in the genomes of mycobacteriophages, which encodes a protein (LysB) with

mycolyl-arabinogalactan esterase activity. Taking in consideration the complex

mycobacterial cell envelope that mycobacteriophages encounter during their life cycle, it

is valuable to evaluate the role of these proteins in lysis. In the present work, we

constructed an Ms6 mutant defective on lysB and showed that Ms6 LysB has an important

role in lysis. In the absence of LysB, lysis still occurs but the newly synthesized phage

particles are deficiently released to the environment. Using cryo-electron microscopy and

tomography to register the changes in the lysis phenotype, we show that at 150 min post-

adsorption, mycobacteria cells are incompletely lysed and phage particles are retained

inside the cell, while cells infected with Ms6wt are completely lysed. Our results confirm

that Ms6 LysB is necessary for an efficient lysis of Mycobacterium smegmatis, acting,

similarly to spanins, in the third step of the lysis process.


75

2.1. Introduction

Bacteriophages, the viruses of bacteria, are key elements for biosphere equilibrium,

playing a fundamental role in bacterial evolution through constant interactions with their

hosts (Rodriguez-Valera et al., 2009; Hatfull et al., 2011). To guarantee their own

survival, double-stranded DNA (dsDNA) phages, which represent more than 95% of

known bacterial viruses (Ackermann et al., 2012), must lyse their hosts. At the end of a

lytic cycle, the new phage particles need to be released into the environment, where new

host bacteria are potentially available for new infection cycles. The main barrier to phage

release is the bacterial cell envelope, and thus, compromising this barrier is the main goal

of the lytic process. To accomplish this goal, dsDNA phages synthesize two essential

lysis proteins, endolysins and holins. Endolysins are enzymes that disrupt the bacterial

cell wall (CW) by cleaving one or more of the five bonds in peptidoglycan (PG). Holins

are small proteins that accumulate in the cytoplasmic membrane (CM) and that, at a

genetically defined time, form holes in this cell membrane allowing the access of active

endolysins to the PG layer or the activation of previously exported endolysins (Young et

al., 2006; Catalão et al., 2013). Phages that infect Gram-positive hosts only require the

synthesis of these two proteins to compromise the bacterial envelope and consequently

for cell burst. However, phages that infect Gram-negative hosts have to face an additional

barrier, the outer membrane (OM). It has been shown recently that disruption of this

barrier is also required for cell lysis (Rajaure et al., 2015). This is achieved by a third

class of lysis proteins named spanins. The best studied spanins are the λ Rz and Rz1

proteins which are an inner membrane and outer membrane protein, respectively. These

two proteins form a complex that spans the entire periplasm mediating the fusion of the

CM with the OM. This results in the elimination of the last barrier to phage release and

consequently, lysis of the host (Berry et al., 2008, 2010, 2012). Spanin genes, which may

encode a sole protein (T1 Gp11) or two subunits like the λ Rz and Rz1 proteins, have

been identified in nearly all phages infecting Gram-negative hosts (Summer et al., 2007;

Krupovič et al., 2008). This indicates that, for phages infecting Gram negative hosts, lysis

is a three-step event where each component of the cell envelope, i.e., CM, CW and OM


76

is sequentially attacked by holins, endolysins and spanins, respectively (Berry et al.,

2012).

Studies of mycobacteriophage Ms6, a phage that infects Mycobacterium smegmatis,

have shown that the lysis cassette composition reflects the complexity of the cell envelope

of its host (Pimentel, 2014). Although mycobacteria are classified as Gram-positive

bacteria, they have a complex cell envelope composed of a CM, similar to other bacterial

CMs (Mamadou et al., 1989; Daffé, 2008), surrounded by a peptidoglycan layer

covalently linked to arabinogalactan (AG) which is in turn esterified to a mycolic acid

(MA), forming the mycolyl arabinogalactan-peptidoglycan (mAGP) complex (Brennan,

2003). The MAs are long fatty acids that constitute the inner leaflet of a true OM. The

outermost leaflet is composed of various glycolipids, including trehalose mono and

dimycolate, phospholipids and species-specific lipids (Hoffmann et al., 2008; Zuber et

al., 2008). Finally, a capsule is composed of proteins, polysaccharides and a few lipids

(Lemassu et al., 1996; Sani et al., 2010). Thus, phages that infect mycobacteria have to

overcome this complex envelope for a successful infective cycle. The Ms6 lysis cassette

is composed of five genes (Garcia et al., 2002) (Figure 2.1). In addition to the holin and

the endolysin functions, Ms6 encodes a chaperone-like protein (Gp1) that is involved in

the delivery of the endolysin to the PG (Catalão et al., 2010, 2011 a) and an additional

lysis protein, Lysin B (LysB), identified as a lipolytic enzyme with the ability to cleave

ester bonds of both short and long fatty acids (Gil et al., 2008). Experiments with

components of the mycobacterial cell envelope showed that Ms6 LysB is a mycolyl-

arabinogalactan esterase that cleaves the ester bond between the mycolic acids and the

arabinogalactan, and this allows the separation of the OM from the CW (Gil et al., 2010).

Analogies can be made between Ms6 LysB and the spanins, where Ms6 LysB functions

to mediate the final step of host cell lysis.

In the present work, we examine the importance of Ms6 LysB in phage lysis and taking

advantage of cryo-electron microscopy (cryo-EM) and tomography (cryo-ET), we

compare the Ms6 wild-type lysis phenotype with that of a Ms6 mutant lacking the lysB


77

gene. We present evidence that absence of LysB in the Ms6 infection cycle results in

incomplete lysis and suggest that the LysB role in lysis parallels that of spanins.

Figure 2.1. Cell envelopes of bacteria (left) and representative lysis cassettes of their

infecting phages (right). (A) Gram-positive bacteria; (B) Gram-negative bacteria; (C)


78

Mycobacteria. The white segments in holin-like genes indicate the number and position

of transmembrane domain coding sequences. Abbreviations: CapGlu, capsular glucan

CM, cytoplasmic membrane; LA, lypoteichoic acid; LAM, lipoarabinomannan; LP,

lipoprotein; LPS, lipopolysaccharide; P, protein; PG, peptidoglycan; PIMs,

phosphatidylinositol mannosides; PLs, phospholipids; PO, porin; Pp, periplasm; TDM,

trehalose dimycolate; TMM, trehalose monomycolate. Adapted from reference Catalão

et al., 2013 with permission.

2.2. Results

2.2.1 Ms6 lysB deletion decreases viral progeny release

To understand how Ms6 LysB contributes to phage-induced lysis, we took advantage

of the Bacteriophage Recombineering of Electroporated DNA (BRED) strategy

(Marinelli et al., 2008) and constructed an Ms6 derivative mutant lacking gene lysB. The

Ms6ΔlysB was able to form plaques on M. smegmatis at equivalent efficiencies to that of

the wild-type (wt); however, a reduction in plaque size produced by the mutant was

observed (Figure 2.2A). In a complementation assay, where LysB production was

provided from plasmid pAG1, the wild-type phenotype was restored, indicating that

plaque size reduction is a consequence of LysB absence.

To test whether this phenotype results from changes in the phage growth parameters,

one-step growth and single-burst experiments were performed. M. smegmatis cells were

infected with Ms6wt or Ms6ΔlysB at a multiplicity of infection (MOI) of one. The one-

step growth curves (Figure 2.2B) obtained for Ms6wt and Ms6ΔlysB show that the latent

period is similar and that LysB has no effect on the lysis timing; however, the number of

infective particles released after Ms6ΔlysB infection was lower than in an Ms6wt

infection. Single-burst experiments performed to compare the viable progeny released

from single cells show that a Ms6wt infection released an average of 147 ± 27 viable

phages per bacterium, while Ms6ΔlysB yielded a reduced burst size of approximately 53

± 14, where the ± values indicate the mean SD of three independent experiments. Again,


79

when LysB was provided in trans, the wt burst size was restored. These results show that,

although Ms6ΔlysB can accomplish lysis of the host cell, the overall process seems to be

less efficient.

Figure 2.2. (A) Phage plaques formed by Ms6 (top) or Ms6ΔlysB (bottom) on a lawn of

M. smegmatis. The plaques formed by Ms6ΔlysB phage are smaller than the ones formed

by the wild-type Ms6; (B) one-step growth curves of Ms6wt (circles) or Ms6ΔlysB


80

(squares) on M. smegmatis mc2155 show a lower number of plaque-forming units (PFU)

released from Ms6ΔlysB infection. Both curves show similar progression up to 90 min

post-adsorption showing no differences in the timing of lysis. T0 marks the end of the

adsorption and start of the one-step experiment. The PFU/mL at t = 0 was used to

normalize PFU/mL of each time point. For each time point, the mean ± SD of four

independent assays is indicated.

2.2.2 Ms6 is trapped in cell debris in absence of LysB

Taking into consideration the observed lysis defect and that: (i) LysB is produced at a

late stage of the infection cycle as other lysis proteins, (ii) this protein is a lipolytic

enzyme that cleaves the linkage of the mycobacterial OM to the mAGP complex; we

hypothesize that the reduced burst size results from a release defect and not from a

reduction in the number of synthesized phage particles. To address this question, we

performed a time course infection assay with either Ms6wt or Ms6ΔlysB and at each time

point the cell pellet was separated from the supernatant and the number of phage particles

in each fraction was determined. As observed in Figure 2.3, at 90 min post-adsorption the

majority of the phage particles are not yet released and the number of PFU in the

supernatant is similar for both phage infections. However, at 180 min post-adsorption, for

the wt phage infection over 90% of phage particles are free in the supernant and only 7%

are in the pelleted fraction, while for the Ms6ΔlysB infection, a remarkable 47% of total

phage progeny is retained in the pellet. These results confirm that the reduced number of

phage particles obtained for the mutant phage, in the single burst experiment, results from

a deficient cell lysis, where part of the newly synthesized virions are trapped in

incompletely lysed cells.


81

Figure 2.3. Distribution of phage particles in the supernatant and pellet of M. smegmatis

infected with Ms6wt or Ms6ΔlysB. Ms6 is trapped in cell debris in absence of LysB. At

the indicated time points, the distribution of phage particles in the pellet and in the

supernatant was determined as a percentage of the total amount of PFU counted in both

fractions. The values indicate the mean ± SD of three independent experiments.

2.2.3 Cryo-EM shows incomplete cell lysis in absence of Ms6 LysB

To prove that the unreleased phage particles remained trapped in incompletely lysed

cells, we used cryo-electron microscopy (cryo-EM) and cryo-electron tomography (cryo-

ET). This method allows us to visualize the host cell lysis and the viral progeny in their

native environment and to examine the lysis behavior of M. smegmatis infected with


82

Ms6wt or Ms6ΔlysB. From each infected M. smegmatis culture, with either Ms6wt or

Ms6ΔlysB, aliquots were plunge frozen on copper grids for cryo-EM assessment.

Figure 2.4 shows collected images of infected cells at 90 and 150 min post-adsorption.

At 90 min, no lysis is yet observed (Figure 2.4 A,C). At 150 min post-adsorption, cells

infected with Ms6wt burst and release almost all the phages (Figure 2.4B), while cells

infected with Ms6ΔlysB show incomplete lysis and many phages are not released (Figure

2.4D). Incompletely lysed cells are still captured up to 240 min post-adsorption with the

mutant phage, while for the wild-type infection only free phage particles and cell debris

are observed.


83

Figure 2.4. Cryo-EM images of M. smegmatis infected with Ms6wt or Ms6ΔlysB. At 90

min post-adsorption, cells infected with Ms6wt (A) or Ms6ΔlysB (C) are still intact and

no difference is observed. At 150 min post-adsorption, the abrupt burst of a cell infected

with Ms6wt is clear (B) while cells infected with Ms6ΔlysB (D) do not lyse abruptly and

deformations in the cell envelope are clearly visible. Scale bar (200 nm).

Cryo-ET data collection was performed on Ms6ΔlysB infected cells. In Figure 2.5A, a

central slice through the 3D tomogram shows phages inside the incompletely lysed cell

and what appears to be lesions throughout the cell envelope. To facilitate the visualization

of the phages and to demonstrate they are inside the cell, segmentations of several 3D

tomographic volumes was performed. With this method, it is possible to render the

structures or regions of interest in the tomogram (Figure 2.5B). It is clear that many

phages are inside the incompletely lysed cell despite the evident deformation of the cell

envelope. It is also clear that most of the CM and PG are absent, while the OM still

remains as a veil surrounding and holding some of the cell content.


84

Figure 2.5. Cryo-electron tomography of M. smegmatis infected with Ms6ΔlysB at 150

min post-adsorption. (A) Slice through the tomogram of an infected cell; (B) segmented

volume of the phage capsids (purple), cell membrane and PG (magenta) and the outer

membrane (green). Scale bar (200 nm).

2.3. Discussion

It is well known that lysis of the bacterial host is the last event of dsDNA phage lytic

cycle, so that the new synthesized phage particles may be released into the environment

and infect new available hosts. Compromising the bacterial cell barriers is a sine qua non

condition to achieve this final step. Although the role of holins and endolysins has long

been well defined, targeting the CM and the CW respectively, the importance of spanins

in lysis has only recently been established (Berry et al., 2012). The best characterized

spanin is that of phage λ, which is composed of two subunits, the Rz and Rz1 proteins,

that, once localized to the inner and outer membranes, respectively interact by the C-

termini of their periplasmic domains to form a complex that spans the entire periplasm

(Summer et al., 2007; Berry et al., 2008). For many years, Rz/Rz1 were considered

auxiliary genes, because under laboratory conditions λ lysis could be achieved in the

absence of these genes, unless the OM was artificially stabilized by the presence of

millimolar concentrations of Ca2+ (Zhang et al., 1999). Recently, Berry et al. have

demonstrated that, in nature, in absence of stabilizing cations, these proteins are required

for λ lysis, as induction of λ lysogens in the absence of a spanin function results in lysis

failure (Berry et al., 2008, 2012). The infection cycle terminates leading to spherical cells

where the CM and PG have been disrupted, but the OM remains intact, indicating that the

latter is an important barrier to lysis. In a λ lytic cycle, the Rz-Rz1 complexes accumulate

in the envelope during the morphogenesis phase. It was suggested that, following PG

disruption by the endolysins, the spanins function by fusing the inner and outer

membrane, this results in outer membrane disruption and consequently cell lysis (Berry

et al., 2008, 2012).


85

This spherical phenotype in the absence of spanins has also been observed in infections

with phages P2 (Markov et al., 2004) and PRD1 (Krupovič et al., 2008). The presence of

Rz/Rz1 equivalents in the lysis cassette, of nearly all bacteriophages that infect Gram-

negative hosts, strengthens the idea that to accomplish lysis, in addition to compromising

the CM and the CW through the action of holins and endolysins, these phages also need

to disrupt the OM (Summer et al., 2007; Berry et al., 2008).

In this work, we show that mycobacteriophage Ms6, a phage that infects the

mycobacterial species, M. smegmatis, in addition to the holin and endolysin functions

also requires an additional lysis protein to overcome the last cell barrier. We provide

evidence that Ms6 LysB parallels the function of spanins.

Mycobacteria, which are members of the Corynebacteriales order, are bacteria that,

despite being classified as Gram-positive, share a complex cell envelope. In addition to a

CM and a CW, they also contain an OM, which is an asymmetrical bilayer where the

inner leaflet, composed of long chain mycolic acids, is linked to the CW through an ester

bond to AG. This peculiar OM confers to mycobacteria their characteristic

impermeability and resistance to therapeutic agents, and as so, is also predicted to be a

barrier to mycobacteriophage-induced lysis (Mamadou et al., 1989; Brennan, 2003;

Daffé, 2008; Hoffmann et al., 2008; Zuber et al., 2008). We have previously shown that

Ms6 LysB is a lipolytic enzyme that cuts the linkage between AG and MA on the

mycolyl-arabinogalactan-peptidoglycan complex (Gil et al., 2008, 2010).

We have observed that, in contrast with Ms6 LysA (Catalão et al., 2011 b) and in general

with phage endolysins, the Ms6 LysB, under our laboratory conditions, is not essential

for the phage life cycle, since the Ms6 derivative mutant lacking gene lysB is viable and

capable of forming plaques in M. smegmatis. We have observed, however, a reduction in

the plaque size. In a one-step assay, we could demonstrate that in absence of LysB there

is a defective phage release at the end of an infection cycle. Indeed, in the single burst

experiment a reduction of 64% in the number of free phage particles per bacterium was

observed. This is in agreement with the reduced plaque size of Ms6ΔlysB, a phenotype


86

that was reverted to the wild-type when LysB was provided in trans. Since Ms6 LysB is

produced during late gene expression, from gene lysB, which is part of the lysis cassette,

a role in host lysis is obviously expected.

A reduced phage release, together with a reduced plaque size was also reported for the

LysB of mycobacteriophage Giles; however, and in contrast to phage Ms6 where absence

of LysB did not affect the timing of lysis, the authors observed that lysis induced by

GilesΔlysB was delayed in 30 min when compared to the Gileswt (Payne et al., 2009,

2012). It is not clear so far how the absence of Giles LysB affects the timing of lysis.

The observation that, at the end of a Ms6ΔlysB infection, 47% of the phages are recovered

from the cell pellet against only 7% in a Ms6wt infection, indicates that the reduced burst

size results from a deficient phage release and not from a reduction in the number of new

synthesized phage particles. In the absence of LysB, phage particles are trapped in

incompletely lysed cells. Cryo-EM and cryo-ET of infections in the absence of LysB

clearly show unreleased phage particles inside cells infected with Ms6ΔlysB, while at the

same time point (150 min post-adsorption) in a Ms6wt infection the cell completely

bursts. The 3D tomogram (Movie S1) also shows deformations of the cell, indicating that

the OM still holds part of the cell content even after disruption of the CM and PG (Figure

2.5B), following holin and endolysin action.

Our results support the notion that the role of Ms6 LysB in lysis equates to that of spanins,

however with different modes of action, since the structure of mycobacteria OM is

completely different from that of Gram-negative bacteria. While spanins function either

as a complex (λ Rz-Rz1) or as a single protein (T1 Gp11) by fusing the CM and OM

(Berry et al., 2013), Ms6 LysB protein functions as an enzyme that detaches the OM from

the CW by cleaving the bond that links these two structures. As a lipolytic enzyme, Ms6

LysB also acts as an esterase on other lipids containing mycolic acids, such as the

trehalose dimycolate (TDM) (Gil et al., 2010), a glycolipid with an important role in M.

tuberculosis pathogenesis (Jackson, 2014). However, it is unknown if cleavage of these

lipids contributes to lysis.


87

The fact that the vast majority of mycobacteriophages sequenced so far encode Ms6 LysB

homologous proteins suggests that they have an important role in nature. This is also true

for other phages that infect members of the mycolata group, a bacterial group that also

contain a layer of mycolic acid-containing lipids in their envelope. Examples are the

Rhodococcus equi phages ReqiDocB7, ReqiPepy6 and ReqiPoco6, which encode Ms6

LysB homologues (Summer et al., 2011; Pimentel, 2014). A huge number of genome

sequences from phages infecting the same bacterial group is available at The

Actinobacteriophage Database (http://phagesdb.org/), and here we can also find genes

from several phages annotated as coding for Lysin B as exemplified by gp24 or gp41

from phages SoilAssassin and Ghobes, respectively, both infecting Gordonae terrae

(Pope et al., 2017). In other cases, although no LysB annotation exists, we could identify

the GXSXG motif common to lipolytic enzymes in the deduced amino acid sequence of

several genes, such as gp54 from the TPA4 phage, a lytic phage that infects Tsukamurella

species (Figure S2.1).

Collectively our results lead to the suggestion that mycobacteriophage-induced lysis is

also a three-step process where holins subvert the cytoplasmic membrane followed by

endolysins targeting the cell wall and LysB proteins disrupting the last barrier to

mycobacteriophage release, the outer membrane.

Our present knowledge of the mechanism of bacteriophage lysis suggests that the

complexity of phage lytic cassettes depends on their hosts. Hosts with a simpler envelope,

like Gram-positive bacteria, require the phage to possess a simple lytic cassette, with

genes encoding proteins targeting the CM and the PG. For bacteria with a more complex

envelope that also contain an OM, degradation of the cell wall is necessary but not

sufficient for lysis and phages need to produce specific proteins to overcome this barrier.

Thus, phages that infect Gram-negative hosts or mycobacteria, in addition to holins and

endolysins, synthesize spanins or lipolytic enzymes, respectively (Figure 2.1).


88

2.4. Materials and methods

2.4.1 Bacterial strains, phages, plasmids and culture conditions

Mycobacteria strains, phages, plasmids and oligonucleotides used in this study are listed

in Table 1. M. smegmatis strains were propagated in 7H9 medium (BD Biosciences) with

shaking or Middlebrook 7H10 (BD Biosciences), supplemented with 0.5% glucose, at

37°C. When appropriate, 1 mM CaCl2 or 15 µg/mL kanamycin was also added to the

media. For induced conditions, cells were grown in 7H9 supplemented with 0.2%

succinate and induced with 0.2% acetamide.

Table 2.1. Bacterial strains, phages, plasmids and oligonucleotides used in this study.

Name Description Source or Reference

Bacteria Mycobacterium

smegmatis mc2155 High-transformation-efficiency mutant of M.

smegmatis ATCC 607

(Snapper et al., 1990)

Bacteriophages

Ms6wt Temperate bacteriophage from M. smegmatis (Portugal et al., 1989)

Ms6ΔlysB 996 bp in-frame deletion of the Ms6 lysB gene This study Plasmids

pJV53 pAG1

Derivative of pLAM12 with Che9c 60 and 61 under control of the acetamidase promoter; Kanr lysB gene cloned into pVVAP

(van Kessel et

al., 2007) This study

pVVAP Mycobacteria shuttle vector carrying the acetamidase promoter; Kanr

(Vissa et al., unpublished)

Oligonucleotides Sequence 5’-3’ a

PrΔlysB

CTCGGCGGAAAAACCCTCCTCGTGGACGCGGTAGCAGAACTGTTGGGCCACTGATAGGAGGCACCCATGCTGACACGTTCATTCTGGATCGACGCCGCCGAGCG

Ms6ΔlysB

PrExtΔlysBFw CGAGATCCTGCGGCAACTGCGCGGATACAACCTCACTGGCTGGCCGCAGCTCGGCGGAAAAACCCTCGTGGACG

Extend Pr∆lysB

PrExtΔlysBRv CCCCGGCGCCGAGGGTGGCGATCGCGGTTTGGGCGAATGTGCGTATGGCACGCTCGGCGGCGTCGATCCAGAATG

Extend Pr∆lysB

PrlysBFw GCGGATCCATGAGCAGAACTGTTGGGCC Includes BamHI site to clone in pVVAP


89

PrlysBRv GGAAGCTTTGTGCGTAGGTAGTCGATG Includes HindIII site to clone in pVVAP

DADA ΔlysB PCRFw

GCGCTAGCAGAACTGTTGGGCCACTGATAG Ms6ΔlysB

DADA Ms6-PCRRv CGTCTCGTACTGCACGTACCGGTTCTTC Ms6ΔlysB

a Restriction sites are underlined.

2.4.2 Construction of Ms6 mutant phage

Construction of Ms6 mutant phage was performed using Bacteriophage

Recombineering of Electroporated DNA (BRED) in M. smegmatis as described

previously (Marinelli et al., 2008; Catalão et al., 2010). Briefly, for deletion of gene lysB

from the Ms6 genome, a 100 bp oligonucleotide (Pr∆lysB), with 50 bp of homology to

either flanking region to be deleted was generated. This fragment was extended by PCR

to a 200 bp dsDNA substrate using two 75 bp extender primers, PrExt∆lysBFw and

PrExt∆lysBRv, sharing 25 bp of homology with either end of the 100-mer. After

purification, using MinElute PCR Purification Kit (QIAGEN), the 200 bp substrate was

co-electroporated with Ms6wt DNA into electrocompetent recombineering cells of M.

smegmatis mc2155:pJV53. Cells were resuspended in 7H9 supplemented with glucose

and CaCl2, incubated at 37 °C for 2 h (prior to lysis) and plated on top agar lawns with

M. smegmatis mc2155. Individual plaques were recovered and eluted in 100 µL of phage

buffer (10 mM Tris-HCl, pH 7.5, 10 mM MgSO4, 68.5 mM NaCl, 1 mM CaCl2), for two

hours at room temperature and analyzed by Deletion Amplification Detection Assay

(DADA)-PCR (Marinelli et al., 2008) with primers DADA ∆lysB-PCRFw/DADA Ms6-

PCRRv to detect lysB deletion. Mixed primary plaques containing both wild-type and

mutant alleles were eluted as described above, and serial dilutions were plated with M.

smegmatis. Individual secondary plaques were screened by DADA-PCR using the same

pair of primers.


90

2.4.3 Plasmid construction

To construct plasmid pAG1, a DNA fragment containing the lysB gene was obtained

by PCR amplification using Ms6 genomic DNA as template, with primers

PrlysBFw/PrlysBRv and Pfu high-fidelity polymerase (Promega®). Primers were

designed in order to have the restriction sites that allow the correct insertion into the

shuttle vector pVVAP (V. Visa and M. McNeil; unpublished). All oligonucleotides used

were purchased from Thermo Scientific (Waltham) and are listed in Table 2.1. DNA

amplification, plasmid isolation and electrophoresis were carried out using standard

techniques (Sambrook et al., 2001). All constructs used in this study were validated and

verified by nucleotide sequencing.

2.4.4 One step growth and single burst experiments

One-step growth curves and burst size determination assays (Adams M. H., 1959) were

adapted to mycobacteria and carried out in exponential growth phase cell cultures

(Catalão et al., 2010). Briefly, 108 M. smegmatis cells were suspended in 1 mL of a phage

suspension at 108 plaque forming units (PFU)/mL. After 50 min of adsorption at 37 °C,

nonadsorbed phages were inactivated with 100 µL of 0.4% H2SO4 for 5 min followed by

neutralization with 100 µL of 0.4% NaOH. The mixture was diluted 1:100 in 7H9 media

and aliquots were taken at intervals of 30 min to quantify the number of phage particles

(Catalão et al., 2010). The obtained results are means of three independent experiments.

A similar procedure was used for burst size determination except that 10 µL of infected

cells were diluted in supplemented 7H9 in order to obtain one infected cell/mL. Then, 50

mL of the infected culture was aliquoted into 1 mL volumes and incubated for 3 h at

37°C. Each sample was plated with 200 µL of M. smegmatis cells and top agar (4 mL) on

7H10 medium and incubated at 37 °C for 24 h. Phage plaques were counted, and the

Poisson distribution of (P(n)) was applied to determine the burst size (BS): P(n) = (e−c ×

cn)/n! (e < 1), where P(n) is the probability of samples having n infected cells, c is the

average number of infected cells per tube, and BS (total plaque count in the 50


91

plates)/(total number of infected cells) (Catalão et al., 2010). The obtained results are

means of three independent experiments.

2.4.5 Determination of the number of phage particles released during the

infection cycle

To determine the number of phage particles released into the supernatant or retained

in cells, M. smegmatis was grown up to an OD600 = 0.5, infected with Ms6wt or Ms6ΔlysB

at a MOI of 1 and incubated at 37 °C for 3 h. Aliquots were taken at 90 min and 180 min

post adsorption and separated by centrifugation into supernatant and pellet fractions. The

pellets were suspended in ice-cold phage buffer and sonicated twice for 5 s, with a 30 s

interval. Each supernatant and the sonicated pellets were serially diluted using phage

buffer and plated on a top agar lawn of M. smegmatis to determine the number of phage

particles. Data represent the mean of three independent experiments.

2.4.6 Cryo-transmission electron microscopy sample preparation, imaging

and image processing

To observe the lysis phenotype of Ms6wt or Ms6ΔlysB, cells were infected as

described above for the one step growth experiment except that the phage input was 100-

fold higher. At each time point, 200 µL aliquots were mixed with 10-nm gold

nanoparticles (Sigma-Aldrich®). The nanoparticles were later used for image alignment

in the 3D tomographic reconstruction process (Mastronarde, 1997; Briken et al., 2004).

Four µL of the pre-mixed samples were applied to TEM grids that were vitrified by rapid

immersion in liquid ethane using a Gatan CryoPlunger3 (Cp3) apparatus (Gatan,

Pleasanton, CA, USA). Cryo-grids were transferred to a Gatan 914 high-tilt holder

maintained at −178 °C. Cryo-specimens were imaged with JEOL JEM-2200FS 200-kV

field emission gun transmission electron microscope (JEOL Ltd., Tokyo, Japan) equipped

with an in-column Omega energy filter (slit width 20 eV), a Gatan US4000 4k × 4k CCD

camera, and a Direct Electron DE-20 direct detector (Direct Electron, LP, San Diego, CA,

USA). Projection images and tilt series were acquired using SerialEM software

(Mastronarde, 2005). Single-axis tilt series were collected over an angular range of −62°


92

to 62°, with a 2° tilt increment using the DE-20 direct detector. The total electron dose

applied to the specimens did not exceed 120 e−/Å2. Tilt series images were acquired at

10,000× nominal magnification (calibrated pixel size of 0.614 nm) with −4 to −8 μm

defocus applied. Tomographic reconstructions were generated with IMOD using the r-

weighted back-projection algorithm (Kremer et al., 1996; Mastronarde, 1997).

2.4.7 Nucleotide sequence accession numbers

The phage genome sequences provided in Figure S2.1 were obtained from GeneBank.

The accession numbers are AF022214 for D29, DQ398047 for PBI1; AF319619 for Ms6,

DQ398042 for Halo, AY129338 for Omega, GU580941 for ReqiPepy6, GU580940 for

ReqiDocB7, KU963246 for SoilAssassin, KX557278 for Ghobes and KR053196 for

TPA4.

2.5. References

Ackermann, H.W.W., Prangishvili, D., (2012) Prokaryote viruses studied by electron

microscopy. Archives of Virology., Vol. 157pp. 1843–1849.

Adams M. H., (1959) Methods of study of bacterial viruses. Bacteriophages. Adams M.

H. Interscience Publishers, Inc., New York, pp. 443–522.










93

Berry, J.D., Rajaure, M., Young, R., (2013) Spanin function requires subunit

homodimerization through intermolecular disulfide bonds. Molecular


Brennan, P.., (2003) Structure, function, and biogenesis of the cell wall of Mycobacterium

tuberculosis. Tuberculosis., 83, 91–97.

Briken, V., Porcelli, S.A., Besra, G.S., Kremer, L., (2004) Mycobacterial

lipoarabinomannan and related lipoglycans: From biogenesis to modulation of the

immune response. Molecular Microbiology.

Catalão, M., Gil, F., Moniz-Pereira, J., Pimentel, M., (2011)a Functional analysis of the

Holin-Like proteins of mycobacteriophage Ms6. Journal of Bacteriology., 193,

2793–2803.




Catalão, M.J., Milho, C., Gil, F., Moniz-Pereira, J., Pimentel, M., (2011)b A second



Catalão, M.J., Gil, F., Moniz-Pereira, J., São-José, C., Pimentel, M., (2013) Diversity in

bacterial lysis systems: bacteriophages show the way. FEMS microbiology

reviews., 37, 554–571.

Daffé, M., (2008) 1 The Global Architecture of the Mycobacterial Cell Envelope. The

Mycobacterial Cell Envelope., pp. 3–11.





Gil, F., Catalão, M.J., Moniz-Pereira, J., Leandro, P., McNeil, M., Pimentel, M., Catalão,





94




Hatfull, G.F., Hendrix, R.W., (2011) Bacteriophages and their genomes. Current Opinion

in Virology., 1, 298–303.





Jackson, M., (2014) The mycobacterial cell envelope-lipids. Cold Spring Harbor

Perspectives in Medicine., 4.

Kremer, J.R., Mastronarde, D.N., McIntosh, J.R., (1996) Computer visualization of three-

dimensional image data using IMOD. J. Struct. Biol., 116, 71–76.

Krupovič, M., Cvirkaitė-Krupovič, V., Bamford, D.H., (2008) Identification and

functional analysis of the Rz/Rz1-like accessory lysis genes in the membrane-

containing bacteriophage PRD1. Molecular Microbiology., 68, 492–503.

Lemassu, A., Ortalo-Magné, A., Bardou, F., Silve, G., Laneélle, M.A., Daffé, M., (1996)

Extracellular and surface-exposed polysaccharides of non-tuberculous

mycobacteria. Microbiology., 142, 1513–1520.

Mamadou, D., Marie-Ange, D., Nicole, G., (1989) The cell envelope of Mycobacterium

smegmatis: cytochemistry and architectural implications. FEMS microbiology

letters., 52, 89–93.




3, e3957.

Markov, D., Christie, G.E., Sauer, B., Calendar, R., Park, T., Young, R., Severinov, K.,

(2004) P2 growth restriction on an rpoC mutant is suppressed by alleles of the Rz1

homolog lysC. Journal of Bacteriology., 186, 4628–4637.

Mastronarde, D.N., (1997) Dual-axis tomography: an approach with alignment methods

that preserve resolution. Journal of structural biology., 120, 343–352.


95

Mastronarde, D.N., (2005) Automated electron microscope tomography using robust

prediction of specimen movements. Journal of Structural Biology., 152, 36–51.






Pope, W.H., Mavrich, T.N., Garlena, R.A., Guerrero-Bustamante, C.A., Jacobs-Sera, D.,

Montgomery, M.T., Russell, D.A., Warner, M.H., Science Education Alliance-

Phage Hunters Advancing Genomics and Evolutionary Science (SEA-PHAGES),

Hatfull, G.F., (2017) Bacteriophages of Gordonia spp. Display a Spectrum of

Diversity and Genetic Relationships. (Losick, R., Ed.)mBio., 8, e01069-17.





of America., 112, 5497–5502.

Rodriguez-Valera, F., Martin-Cuadrado, A.B., Rodriguez-Brito, B., Pasić, L., Thingstad,

T.F., Rohwer, F., Mira, A., (2009) Explaining microbial population genomics

through phage predation. Nature reviews. Microbiology., 7, 828–836.

Sambrook, J., Russell, D.W., (2001) Molecular Cloning - Sambrook & Russel. Human

Mutation., Vol. 18.

Sani, M., Houben, E.N.G., Geurtsen, J., Pierson, J., de Punder, K., van Zon, M., Wever,

B., Piersma, S.R., Jiménez, C.R., Daffé, M., Appelmelk, B.J., Bitter, W., van der

Wel, N., Peters, P.J., Van Wel, N. Der, Peters, P.J., (2010) Direct visualization by

Cryo-EM of the mycobacterial capsular layer: A labile structure containing ESX-

1-secreted proteins. (Ramakrishnan, L., Ed.)PLoS Pathogens., 6, e1000794.

Snapper, S.B., Melton, R.E., Mustafa, S., Kieser, T., Jr, W.R.J., (1990) Isolation and

characterization of efficient plasmid transformation mutants of Mycobacterium

smegmatis. Molecular Microbiology., 4, 1911–1919.


96



Biology., 373, 1098–1112.






van Kessel, J.C., Hatfull, G.F., (2007) Recombineering in Mycobacterium tuberculosis.

Nature Methods., 4, 147–152.

Young, R., Wang, I.N., (2006) Phage lysis. In: Calendar, R. (ed.), The Bacteriophages.

2nd edn. Oxford University Press, New York, NY, pp. 104–125.

Zhang, N., Young, R., (1999) Complementation and characterization of the nested Rz and

Rz1 reading frames in the genome of bacteriophage λ. Molecular and General

Genetics., 262, 659–667.





97

2.6. Supplementary materials


98

Figure S2.1 Comparison of lysB genes from phages infecting members of the Mycolata

group. (A) Illustrated are representatives of mycobacteriophages and phages infecting

Rhodococcus equi, Gordoniae terrae and Tsukamurella spp.. The genes marked with a

white segment indicate genes not previously assigned as holins but having predicted

transmembrane segments. (B) Alignment of Ms6 LysB and putative LysB protein

homologues. The conserved pentapeptide (G/A-X-S-X-G) is highlighted on a grey

background. Numbers refer to the amino acid positions.

Additional supplementary material published on Gigante, A.M., et al., (2017) available

online at:

http://www.mdpi.com/1999-4915/9/11/343/s1

http://www.mdpi.com/1999-4915/9/11/343/s1

CHAPTER 3.

THE N-TERMINAL OF

MYCOBACTERIOPHAGE MS6 LYSB

REVEALS PEPTIDOGLYCAN BINDING

CAPACITY

Manuscript in preparation

The N-terminal of mycobacteriophage Ms6 LysB reveals peptidoglycan binding capacity

Gigante, A., Catalão M.J., Olivença F., Moniz-Pereira J, Filipe, S., Pimentel M.

The N-terminal of mycobacteriophage Ms6 LysB reveals

peptidoglycan binding capacity

Gigante, A1., Catalão MJ1, Olivença F1, Moniz-Pereira J1, Filipe S2,3, Pimentel M1



2UCIBIO-REQUIMTE, Departamento de Ciências da Vida, Faculdade de Ciências e Tecnologia,

Universidade Nova de Lisboa, Caparica, Portugal

3Laboratory of Bacterial Cell Surfaces and Pathogenesis, Instituto de Tecnologia Química e Biológica

António Xavier, Universidade Nova de Lisboa, Oeiras, Portugal

Running Title: Ms6 LysB N-terminus PG binding capacity

Keywords: mycobacteriophage; Ms6; LysB; phage lysis, mycobacteria, peptidoglycan binding


3. Ms6 LysB N-terminus PG binding capacity

103

Acknowledgments

This work was supported by Fundação para a Ciência e Tecnologia (FCT-MCES,

Portugal) Grant PTDC/IMI-MIC/0694/2012 to MP; AG (SFRH/BD/87685/2012) is a

recipient PhD fellowship from FCT-MCES, Portugal.






105

Abstract

Double-stranded DNA bacteriophages end a lytic cycle by disrupting the cell envelope,

which allows the release of the virion progeny. Each phage must synthesize lysis proteins

that target each cell barrier to phage release. In addition to holins that permeabilize the

cytoplasmic membrane and endolysins that disrupt the peptidoglycan,

mycobacteriophages synthesize a specific lysis protein, LysB, to detach the outer

membrane from the complex cell wall of mycobacteria. The family of LysB proteins is

highly diverse with many proteins presenting an extended N-terminus. The N-terminal

region of mycobacteriophage Ms6 LysB showed structural similarity to the

peptidoglycan-binding domain of the φKZ endolysin. A fusion of this region with

enhanced green fluorescent protein (Ms6LysBPGBD-EGFP) was shown to bind to

Mycobacterium smegmatis, Mycobacterium vaccae, Mycobacterium bovis BGC and

Mycobacterium tuberculosis H37Ra cells pretreated with SDS or Ms6 LysB. In pulldown

assays, we demonstrate that Ms6 LysB and Ms6LysBPGBD-EGFP, bind to purified

peptidoglycan of M. smegmatis, Escherichia coli, Pseudomonas aeruginosa and Bacillus

subtilis, demonstrating affinity to the A1γ chemotype. An infection with an Ms6 mutant

producing a truncated version of LysB lacking the first 90 amino acids resulted in an

abrupt lysis. These results clearly demonstrate that the N-terminus of Ms6 LysB binds to

the peptidoglycan.


107

3.1. Introduction

Bacteriophages, also known simply as phages, are viruses that are able to propagate

by infecting bacteria. The infection process used by double-stranded DNA phages begins

with the binding of the phage particle to a receptor present at the bacterial surface,

followed by injection of the phage DNA into the bacterial host. Once in the cytoplasm

many phages may undergo a lytic cycle and after replication of the phage DNA, and

synthesis of the virion components, they are assembled to form new phage particles. In

the last step, phages need to lyse the host and release the viral progeny to start a new

infection cycle (Young et al., 2006). To achieve this, phages need to overcome the

barriers of the bacterial envelope. The lytic system employed by dsDNA phages depends

on at least two proteins: an endolysin and a holin (Young et al., 2006). Holins are

membrane proteins that permeabilize the cytoplasmic membrane (CM), determining the

timing of lysis. Endolysins are enzymes that cleave covalent bonds in the peptidoglycan

(PG) compromising the integrity of the cell wall (Young et al., 2006; Young, 2014). The

recent reports on phage-induced lysis make it clear that phages that infect Gram-negative

hosts require additional lysis proteins to compromise the last barrier to phage release, the

outer membrane (OM) (Berry et al., 2012). These proteins are named spanins and may

exert their function as a complex formed by two spanin subunits, the i-spanin which

integrates in the CM and the o-spanin, a lipoprotein that anchors to the OM, or as a unique

protein, the u-spanin able to anchor to both membranes (Summer et al., 2007; Young,

2014). As a complex or as a sole protein, these proteins disrupt the OM, by a topological

mechanism (Berry et al., 2008; Rajaure et al., 2015).

The best studied spanins are the Rz and Rz1 proteins of bacteriophage which are

translated from 2 genes with a particular genetic architecture. Rz1 is embedded in the +1

reading frame of Rz and codes for a lipoprotein, while Rz positioned downstream of R

gene (encoding the endolysin) codes for an integral type II membrane protein. These two

proteins are localized to the membranes during the morphogenesis period of an infection

cycle, and interact with each other through their C-terminus forming a complex that spans

the entire periplasm(Kedzierska et al., 1996; Berry et al., 2008). It has been proposed that

phage induced lysis in Gram-negative hosts is a three step event in which holins


108

accumulated in the CM, form holes at a genetically defined time, permeabilizing the

membrane, which consequently allows the endolysin to cleave specific bonds in the PG.

Disruption of the PG meshwork induces conformational changes in the spanins, which

brings the two membranes into close proximity resulting in the fusion of the CM and OM

(Berry et al., 2008, 2010; Young, 2013) eliminating the last barrier of the cell envelope.

Phages that infect mycobacteria have to face a more complex cell envelope where the

presence of an outer membrane also constitutes a third barrier in the lysis process.

However, the composition of the mycobacteria OM is completely different from the OM

of Gram-negative bacteria. It is a bilayer with an inner leaflet mainly composed by

mycolic acids, and an outer leaflet rich in several free (non-covalently bound) lipids

(glycolipids, phospholipids and species specific lipids) (Hoffmann et al., 2008; Zuber et

al., 2008; Chiaradia et al., 2017). This OM or mycomembrane is covalently linked to

arabinogalactan which in turn is covalently linked to peptidoglycan via an

arabinogalactan network (Jackson, 2014).

Until the time of this writing the complete genome sequence of more than 340

mycobacteriophages is available in GenBank, with most genomes presenting in their lysis

cassette a gene, lysB that codes for a lipolytic enzyme. The work of Gil et al. (2008, 2010)

and Payne et al. (2009) have shown that LysB proteins produced by phage Ms6 and Giles

cleave the ester bond that link the OM to the PG-AG complex. This suggests that these

phages also lyse their hosts in three steps. Studies about lysis induced by

mycobacteriophages are scarce, being the existent ones centralized in

mycobacteriophages Ms6, Giles and D29 (Payne et al., 2009; Pimentel, 2014).

Like the majority of mycobacteriophage lysB genes, the Ms6 lysB is located

downstream of lysA (endolysin), however a multiple sequence alignment of LysB proteins

shows they are highly diverse, with some members showing as low as <20% identity to

each other. However, all have in common with the Ms6 LysB, the motif G-X-S-X-G,

characteristic of lipolytic enzymes (Gil et al., 2008; Payne et al., 2009).

We have previously reported the lipolytic activity of Ms6 LysB (Gil et al., 2008) and

identified the mycobacteria OM as its main target since it cleaves the ester bond between


109

the mycolic acids and arabinogalactan (Gil et al., 2010). Recently we showed that Ms6

LysB is essential for an efficient release of the new progeny virions at the end of a lytic

cycle (Gigante et al., 2017). In the absence of LysB, Ms6 particles remain trapped inside

incomplete lysed cells, which results in a non-profit infection cycle with a 64% reduction

of the burst size.

Bioinformatic analysis has predicted a peptidoglycan binding domain at the N-

terminus of some LysB proteins (Payne et al., 2009; Henry et al., 2011). In the present

work we investigate the relevance of the N-terminus region of Ms6 LysB to the lysis

process and experimentally determined the ability to bind peptidoglycan. In view of the

current knowledge of phage-induced lysis by bacteriophage λ we discuss the role of the

Ms6 LysB peptidoglycan binding domain in the context of mycobacteria lysis.

3.2. Results

3.2.1 Sequence analysis of Ms6 LysB reveals a putative PGBD

Previous studies have shown that Ms6 LysB homologues encoded by

mycobacteriophages belonging to diverse clusters, although sharing the conserved

pentapeptide G-X-S-X-G, characteristic of enzymes with lipolytic activity, are highly

diverse (Payne et al., 2009; Henry et al., 2011). Protein length may vary between 244-

458 amino acids and many LysB proteins have an extra N-terminus, 74-90 residues in

length that is absent in several other LysB proteins (Fig. 3.1A). This diversity is also

extended to the structural level since 32 different types of folds were identified in a CATH

analysis for 72 LysB proteins (Henry et al., 2011).

Analysis of Ms6 LysB by HHPred (Söding et al., 2005; Hildebrand et al., 2009)

predicts two regions, one starting at amino acid 91, containing the catalytic domain,

showed high similarity with the structure of mycobacteriophage D29 LysB (3HC7_A)

amino acids 1-253, with an e-value of 1.2e-39. This region includes a predicted cutinase

motif (120-208) (pfam01083) identified through the MotifFinder tool

(https://www.genome.jp/tools/motif/), which is in agreement with the previously

identified lipolytic activity of this protein (Gil et al., 2008). Residues 1-90 are part of the


110

mentioned extra N-terminus. Although for some LysB homologues a peptidoglycan-

binding motif (PF01471) was predicted within this region (Fruitloop_30, JAMaL_44,

Che9c_26, TM4_30) (Henry et al., 2011; Payne et al., 2012), no such motif was predicted

for the Ms6 LysB. Of note is the fact that Fruitloop_30 shares with the amino acid

sequence of Ms6 LysB an identity of 98% and has the same CATH domain (2mprA00)

(Henry et al., 2011).


111

Figure 3.1 In silico analysis of LysB proteins. (A) Modular structure of KZ144 and

representative mycobacteriophage’s LysB proteins. Regions with structural similarity to

the KZ144 PGBD are shown in light grey blocks and a predicted catalytic region similar

to D29 LysB in dark grey. φKZ gp144 lytic transglycosylase domain is marked with a

strip pattern. The predicted PGBD in Fruitloop_gp30 is marked with a dot pattern. Each

phage cluster is indicated by the letters on the left. The data shown are the result of a

HHpred analysis for each depicted mycobacteriophage LysB. Predicted cutinase domains

with MOTIF Finder are indicated within the LysBs catalytic region. (B) Alignment

between the N-terminus of Ms6 LysB and the PGBD of φKZ endolysin as obtained from

HHPred.

The only experimentally determined structure of a LysB protein is that of

mycobacteriophage D29, and unfortunately, this shorter protein lacks this extra N-

terminus, consequently there is no structural information available for this region.

Interestingly, inspection of the Ms6 LysB N-terminus (not present in D29 LysB) showed

structural similarity with the peptidoglycan-binding domain (PGBD) of the Pseudomonas

aeruginosa bacteriophage φKZ endolysin (KZ144) 3BKH_A, with an e-value of 9.7e-05

(Fig. 3.1B). KZ144 is a lytic transglycosylase containing a PGBD (Pfam01471) at the N-

terminus for which the peptidoglycan binding capacity was experimentally demonstrated

(Briers et al., 2007). In addition, a Blastp search identified a peptidoglycan binding

protein of Mycobacterium fortuitum (WP_064866887) with similarity to Ms6 LysB (49%

sequence identity). Based on these observations it is tempting to hypothesize that the N-

terminus of Ms6 LysB encompasses a peptidoglycan binding domain.

3.2.2 The Ms6 LysB N- terminus binds to mycobacterial cells

To evaluate the binding capacity of the Ms6 LysB N-terminal region, we first

constructed a recombinant protein by fusing the region encompassing amino acids 1-90

of Ms6 LysB to the Enhanced Green Fluorescence Protein (EGFP) and tested the ability

of this protein (Ms6LysBPGBD-EGFP) to bind to mycobacteria cells. Since in

mycobacteria the PG is not exposed due to the presence of an outer membrane, in an

attempt to reach the PG, intact cells were first treated with 1% SDS for 1h, washed with

PBS and subsequently incubated with Ms6LysBPGBD-EGFP or EGFP (negative


112

control). Due to the lipid nature of the mycobacterial outer membrane, treatment with

detergents disturbs the membrane structure and apparently dissolves extractable lipids,

which results in significant changes in permeability (Hoffmann et al., 2008).

Mycobacteria were shown to be resistant to high concentrations of SDS (Manganelli et

al., 1999, Yamamura & Harayama, 2007). Observation of treated cells by fluorescence

microscopy clearly shows that the Ms6LysBPGBD-EGFP labels the cell surface of M.

smegmatis, M. vaccae, M. bovis BCG and M. tuberculosis H37Ra (Fig. 3.2A), resulting

in a bright fluorescence of the entire cell surface, whereas EGFP alone did not label the

cells (Fig. S3.2).

Since Ms6 LysB hydrolyses mycolic acids-containing lipids and the main target is the

link between the OM and the PG-AG, in an attempt to detach the mycobacteria outer

membrane we also performed an experiment where we have pretreated M. smegmatis

cells with the full-length Ms6 LysB protein. Prior to the incubation with the fluorescent

proteins, M. smegmatis cells were incubated with Ms6 LysB for 45 min (see methods for

details). Again, the Ms6LysBPGBD-EGFP was able to bind M. smegmatis cells (Fig.

3.2B) while no fluorescence was observed with the control EGFP (Fig. S3.2). This data

indicate that the Ms6 LysB N-terminus has indeed a role in the binding. No binding was

observed to any untreated cells, indicating that the binding target is not surface exposed.

Importantly, our data also shows the ability of Ms6 LysB to disturb the mycobacteria

surface, allowing the access of the fusion protein to its target. However, the cells are not

homogeneously fluorescent as observed for the cells pretreated with SDS. A possible

explanation for this irregular binding is that the full-length Ms6 LysB also contains the

predicted PGBD N-terminus, which may be competing with Ms6LysBPGBD-EGFP for

the same target.


113

Figure 3.2 – Cell binding capacity of Ms6LysBPGBD-EGFP. (A): M. smegmatis (I), M.

vaccae (II), M. bovis BCG (III) and M. tuberculosis H37Ra (IV) were treated with 1%

SDS prior to incubation with the fusion protein. (B): M. smegmatis cells pretreated with

the full-length Ms6 LysB. Representative images of each strain are shown in phase-

contrast (left) and fluorescence microscopy (right). Exposure time of 100 ms; scale bar 2

µm.


114

3.2.3 Ms6 LysB N- terminus binds to peptidoglycan

To specify that PG is the substrate, we evaluated the binding capacity of Ms6LysB N-

terminal region in a pulldown assay using purified PG from M. smegmatis. 300 μg/ml of

Ms6LysBPGBD-EGFP were added to the pure PG and incubated for 45 min at room

temperature. After centrifugation and washing steps the pellet and supernatant fractions

were analysed by SDS-PAGE. In this assay, proteins that bind to the insoluble PG will

co-precipitate with PG and will be found in the pellet fraction. Comparing the amounts

of protein in each fraction, it is clear that Ms6LysBPGBD-EGFP was pulled down by PG

(Fig. 3.3A). In this experiment, bovine serum albumin (BSA), a protein that does not bind

to PG, was added to the reaction mix, as a control for unspecific binding, to ensure that

the PG pellet was adequately washed, and no unbound protein remains in the pellet

fraction. As expected the BSA band is only detected in the supernatant fraction indicating

that only proteins that have ability to bind PG are in the pellet. To rule out the possibility

of a nonspecific binding, we performed a parallel assay with EGFP, where the PGBD is

not present and, as expected, the protein is only detected in the supernatant fraction (Fig.

3.3A). In addition, to ensure that the presence of Ms6LysBPGBD-EGFP in the pellet

fraction is not a result of protein precipitation, a negative control without PG was

performed in the same conditions. In this case the fusion protein is totally in the

supernatant fraction (Fig. 3.3C). The same ability to be pulled down by PG is observed

with the full-length Ms6 LysB (Fig. 3.3A) where the PGBD is present, strengthen the idea

that the N-terminus of Ms6 LysB has peptidoglycan binding capacity. The amount of

protein that remains free and appears in the supernatant fraction is a result of the excess

of protein used in the experimental setup. By using 0.3 mg/mL of the tested protein we

ensure not only consistency between assays but guarantee that the protein saturates all the

possible binding regions in the PGN, therefore we can assume that the protein in the pellet

fraction is the maximum amount possible that 20 µg of PG can pulldown.

To determine the specificity of PG binding we have also performed pull down assays

exactly in the same conditions of the described above, using PG from the Gram-positive

B. subtilis, S. aureus and S. pneumoniae and from Gram-negative E. coli and P.

aeruginosa. As observed in Fig. 3.3A Ms6LysBPGBD-EGFP or Ms6 LysB were able to


115

bind to the PG of E. coli, P. aeruginosa and B. subtilis, although each PG pulled down

different amounts of protein, which may suggest that the binding affinity is not the same

for all the PG tested. Ms6LysBPGBD-EGFP and Ms6 LysB showed similar PG binding

affinities, with both proteins showing higher affinity for M. smegmatis PG, 67% and 65%

respectively are in the pellet fraction. Over 50% of each protein pulled down by PG of E.

coli or P. aeruginosa and about 35% were pelleted with the Bacillus subtilis PG (Fig

3.3B). No binding was detected for the PG of S. aureus or S. pneumoniae (Fig. 3.4). Of

note is the fact that the binding capacity is observed in bacteria species belonging to the

same PG chemotype. PG of M. smegmatis, E. coli, P. aeruginosa and B. subtillis belong

to A1γ chemotype (a 3→4 direct cross-linkage and meso-2,6-diaminopimelic acid at the

third position of the peptide stem) whereas S. aureus and S. pneumoniae belong to

chemotype A3γ (cross-linkage by interpeptide bridges and L-Lysin at the third position)

(Schleifer et al., 1972; Vollmer et al., 2008 a). These results suggest that a direct cross

linkage and/or the presence of meso-DAP is determinant for Ms6 LysB binding. Although

the PGs that pulled down LysB belong to the same chemotype their glycan strands present

different modifications. B. subtillis strains contain N-deacetylated GlcNAc and MurNAc

residues which may account for a reduced binding as a fully N-acetylated peptidoglycan

is required for the binding of the bacteriophages φKZ and EL endolysins (Briers et al.,

2007). N-glycolylation of the muramic acid is one of the hallmarks of mycobacteria,

however the role of the glycolyl residue is unknown and how this glycan modifications

influences the binding affinity of Ms6 LysB remains to be determined.

As expected for the EGFP purified protein, PG binding was not observed. These results

clearly demonstrate that the Ms6 LysB binding to the PG is dependent on the presence of

the PGBD.


116

Figure 3.3 Pulldown assays of full length Ms6 LysB, Ms6LysBPGBD-EGFP and EGFP

with purified peptidoglycan from: (A) M. smegmatis, E. coli, P. aeruginosa and B.

subtilis. Pellet and supernatant fractions were analyzed by SDS-PAGE followed by

Comassie-blue staining. BSA was used as a control and is only present in the supernatant

fractions. (B) Quantification of the pulldown assay. The gels represented in Panel A were

scanned and band intensities were quantified by densitometry. ImageJ2 software was used

to compare the density of each band. The percentage of bound protein was calculated

using the estimated amount of protein in the pellet fraction (bound) divided by the sum


117

of the estimated protein amount on the pellet (bound) and supernatant (unbound)

fractions. BSA was used as a loading-control (2.5 μg/lane) of all samples and the

corresponding gel band used as a reference for density proportion calculations.

Ms6LysBPGBD-EGFP (dark grey), Ms6 LysB (light grey). (C) As negative control,

Ms6LysBPGBD-EGFP was incubated without peptidoglycan (− PG).

Figure 3.4 Pulldown assays of full length Ms6 LysB with purified peptidoglycan from S.

pneumoniae and S. aureus. Pellet and supernatant fractions were analyzed by SDS-PAGE

followed by Comassie-blue staining. BSA was used as a control and is only present in the

supernatant fractions.


118

3.2.4 Deletion of PGBD from LysB results in a faster rise period

Since this PGBD is in a module separated from the catalytic domain, we investigated

the contribution of this N-terminus region to the phage lysis phenotype. Taking advantage

of the Bacteriophage Recombineering of Electroporated DNA (BRED) technology

(Marinelli et al., 2008), we constructed an Ms6 mutant (Ms6lysBΔPGBD) lacking the

sequence coding for this domain (aa 1-90). This mutant phage produced phage plaques

on M. smegmatis lawn with similar morphology to that of the wt, in contrast with the

Ms6ΔlysB, a mutant that do not produce LysB, forming plaques with a reduced size

(Gigante et al., 2017) (Fig. 3.5).

Figure 3.5. Phage plaque morphology of Ms6lysBΔPGBD (I), Ms6ΔlysB (II) and Ms6wt

(III) on M. smegmatis mc2155 top agar using standard 100mm x 15mm petri dishes (A),


119

and 2x zoom of each highlighted area (B). The mutant Ms6lysBΔPGBD forms slightly

larger plaques than the Ms6wt, while Ms6ΔlysB, in turn, forms the smallest plaques. The

pictures were taken at 24h post infection.

To understand how important is the LysB PGBD to phage lysis, one-step growth

experiments were performed with Ms6wt, Ms6lysBΔPGBD and Ms6ΔlysB. M.

smegmatis cells were infected with Ms6wt, Ms6lysBΔPGBD or Ms6ΔlysB at a

multiplicity of infection (MOI) of one. All phages showed a latent period of 140 min,

consisting of the initial 50 min of adsorption and an additional 90 min as shown in Fig.

3.6, indicating that no changes in the timing of lysis occur. However, the rise period

observed with Ms6lysBΔPGBD is shorter than that of Ms6wt or Ms6ΔlysB. Both Ms6wt

and Ms6ΔlysB showed a slow rise period of 90 min, while Ms6lysBΔPGBD presented a

rise period of 60 min, i.e. the maximum number of released phage particles is achieved at

150 min post-adsorption, in contrast with 180 min for the Ms6wt or Ms6ΔlysB.

Figure 3.6. One-step growth curves of Ms6lysBΔPGBD (triangles), Ms6wt (circles) or

Ms6ΔlysB (squares) on M. smegmatis mc2155. Ms6ΔPGBD has a quick increase in the

number of plaque-forming units (PFU), when compared to a Ms6wt or Ms6ΔlysB

infection. Both curves show similar progression up to 90 min post-adsorption showing no

differences in the timing of lysis. T0 marks the end of the adsorption and start of the one-

step experiment. The PFU/mL at t = 0 was used to normalize PFU/mL of each time point.

For each time point, the mean ± SD of three independent assays is indicated.

Ms6lysBΔPGBD

Ms6wt

Ms6ΔlysB


120

3.3. Discussion

Bacteriophages are the most abundant biological entities on Earth, playing key roles

in the biology of bacteria. For their own survival they have to maintain a strict/close

interaction with their hosts, which for the most frequently isolated phages from nature

(the tailed bacteriophages) leads to bacterial lysis. This is an important characteristic of

bacteriophages, which is being explored as anti-bacterials to fight pathogenic bacteria

particularly antibiotic resistant strains. It is thus of huge importance to understand the

mechanism of phage-induced lysis. To be released at the end of a lytic cycle, phages have

to overcome the cell envelope barriers. It was generally thought that degradation of the

PG was sufficient for cell burst, which will allow the spread of the new synthesized phage

particles to the environment. Although this is true for phages infecting Gram-positive

bacteria it has recently become clear that in Gram-negative hosts the OM is an important

barrier and that its disruption is required for phage lysis. This function is achieved by a

third class of lysis proteins, the spanins (Berry et al., 2012).

The studies performed by Ry Young and collaborators gave a huge contribution to the

comprehension of the role of each phage lysis protein in disruption of the bacterial cell

envelope. It is now clear that each barrier must be sequentially eliminated, beginning with

permeabilization of the inner membrane by the holins, followed by breaking of the

peptidoglycan mesh by the endolysins and in Gram-negative hosts, the last barrier, the

OM, is disrupted by the spanins (Berry et al., 2012, Young, 2014).

In mycobacteriophages, the players in mycobacteria lysis have already been identified

(Garcia et al., 2002). Distinct from other phage lysis cassettes is the presence of gene

lysB, which encodes a lipolytic enzyme with mycolyl-arabinogalactan esterase activity

and whose function is elimination of the lipid rich OM of mycobacteria (Gil et al., 2008,

2010).

In this work we show that the mycobacteriophage Ms6 LysB has a modular

architecture comprising an N-terminal region with structural similarity to the N-terminal


121

peptidoglycan binding domain of phage φKZ endolysin and a predicted C-terminal

catalytic module with structural similarity to the LysB protein of mycobacteriophage

D29. Here we experimentally demonstrate the peptidoglycan binding capacity of Ms6

LysB to PG. The PGBD either present on the full-length Ms6 LysB or in the fusion protein

Ms6LysBPGBD-EGFP allows the binding to the PG of P. aeruginosa, E. coli, B. subtillis

and M. smegmatis, all sharing a A1γ peptidoglycan chemotype, suggesting a common

structure for the PGBD. However, our data suggests that the binding affinity is not

identical in all tested bacteria, with M. smegmatis PG showing the highest percentage of

binding (~65%). The glycan strands of the bacterial peptidoglycan, made of alternating

β-1,4-linked GlcNAc and MurNAc residues, are invariably modified or linked to other

cell wall polymers. The biological role of such modifications (deacetylation, O-

acetylation and N-glycolylation) may vary according to the type of modification.

Deacetylated sugars in the PG strands as well as O-acetylated PG, present an increased

resistance to the activity of lysozyme, a muramidase that cleaves the β-1,4-glycosidic

linkage between MurNAc and GlcNAc (Vollmer et al., 2008 b). This decrease in activity

was correlated to a decrease in substrate binding (Vocadlo et al., 2001; Vollmer et al.,

2008 a). This suggests that chemical modifications of the PG may account for the

observed differences in the binding of Ms6 LysB and Ms6LysBPGBD-EGFP.

Furthermore, MurNAc N-glycolylation, a modification characteristic of the

Mycobacterium genera, was also suggested to have a role in protection against lysozyme

(Raymond et al., 2005; Vollmer, 2008). Although the role of the chemical modifications

to PG binding was not investigated in this work it is reasonable to hypothesize that

modifications on the glycan strands may account for the observed differences in the

binding of Ms6 LysB and Ms6LysBPGBD-EGFP. Nevertheless, our results clearly

demonstrate that the N-terminus of Ms6 LysB is responsible for the binding of the protein

to PG. Ms6 LysB is a late protein that acts in the last step of phage lysis. Similarly to what

happens with phage λ, the Ms6 holins, endolysins and LysB must also act sequentially

with LysB disrupting the OM of mycobacteria after disruption of the PG meshwork by

the Ms6 endolysins, which exert their function after holin trigger. In bacteriophage λ the

spanin subunits Rz and Rz1 have recognized sequence signals to anchor them to the inner

and outer membranes respectively (Hanych et al., 1993; Kedzierska et al., 1996) where

they accumulate and form a complex through interaction of their C-terminus (Berry et al.,


122

2008). While the PG meshwork is intact, spanins are trapped and disruption of OM is

inhibited. Once the endolysin disrupt the PG, Rz and Rz1 undergo conformational

changes that allow the fusion of the IM and OM, removing consequently the OM barrier

to phage release (Berry et al., 2012; Rajaure et al., 2015). Making a parallel it is

reasonable to think that the Ms6 LysB N-terminus could serve to anchor the LysB,

positioning it close to its target, keeping it inactive until disruption of the PG, which

would release the protein and consequently cleave the link to the OM. This means that

LysB would be positioned in the extracytoplasmic environment before holin triggers.

However, in contrast to spanins, no export signals were predicted for the Ms6 LysB and

how it accesses its target is unknown.

The question that arises is what the advantage for the phage fitness of LysB binding to

PG, since deletion of the PGBD results in a one-step curve with a shorter rise period,

meaning that the binding to the PG is delaying the full release of the viral progeny. Once

lysis begins the mutant phage release is more abrupt than in wild-type infection or when

the full length LysB is absent (Fig. 3.6). Many Ms6 LysB homologues encoded in

mycobacteriophage genomes do not contain this N-terminal region as exemplified by the

D29 LysB (Fig. 3.1) and comparison of the phage biological parameters of phages

encoding LysB with and without the N-terminal regions would give additional clues on

the importance of this region to the lysis phenotype. In addition, how Ms6 LysB and

homologues lacking this PGBD reach their target is a matter that deserves future

investigation.

Of note is the fact that Ms6 LysB has proven to disturb the mycobacteria cell envelope

when added externally as treatment of intact cells with the purified protein parallels the

effect of SDS, allowing the binding of the fusion protein Ms6LysBPGBD-EGFP. This is

consistent with the fact that Ms6 LysB, in addition to function as a mycolyl-

arabinogalactan esterase is also able to hydrolase OM membrane containing lipids (Gil et

al., 2010) which would disturb the OM and somehow allow the binding to PG. This means

that LysB proteins may be explored as tools to disturb the mycobacteria OM.

To our knowledge this is the first work demonstrating the ability of a

mycobacteriophage lysis protein LysB to bind to peptidoglycan.


123

3.4. Materials and Methods

3.4.1 Bacterial strains, plasmids, bacteriophages and culture conditions

Bacterial strains, plasmids and bacteriophages used in this study are listed in Table

3.1. E.coli, Staphylococcus aureus, Streptococcus pneumoniae, Bacillus subtilis and

Pseudomonas aeruginosa were grown at 37ºC in Luria-Bertani (LB) broth or agar when

appropriate. For the growth of E. coli recombinant strains media were supplemented with

100 µg/ml ampicillin. M. smegmatis mc2155, M. tuberculosis H37Ra, M. vaccae and M.

bovis BCG were grown at 37ºC in Middlebrook 7H9 or 7H10 (Becton Dickinson)

supplemented with 0.5% (w/v) glucose for M. smegmatis and M. vaccae or OADC

(Becton Dickinson) and 0.05% (v/v) Tween 80 for M. tuberculosis and M. bovis BCG.

For the growth of M. smegmatis recombineering strain carrying pJV53 (van Kessel and

Hatfull, 2007), media was supplemented with 15 μg/ml kanamycin.

3.4.2 General DNA techniques and cloning procedures

DNA manipulations and electrophoresis were carried out using standard techniques

(Sambrook et al., 2001). DNA fragments were amplified by PCR using Pfu polymerase

(Promega). E. coli and M. smegmatis mc2155 cells were transformed as described

previously (Snapper et al., 1990; Sambrook and Russell, 2001). Restrictions enzymes and

T4 DNA ligase were purchased from ThermoScientific and New England Biolabs

respectively and were used in accordance to the manufacturers’ instructions. All

oligonucleotides used are listed in table S1 and were purchased from Thermo Scientific.

The oligonucleotides were designed in order to have restriction sites that allow insertion

into the appropriate plasmids.

3.4.3 Plasmid construction

DNA fragments were amplified by PCR using Pfu high-fidelity polymerase

(Promega). All oligonucleotides, listed in Table S3.1, were purchased from Thermo

Scientific and were designed in order to have the restriction sites that allow the correct


124

insertion into the vector. Restrictions enzymes and T4 DNA ligase were purchased from

ThermoScientific and New England Biolabs respectively and were used in accordance to

the manufacturers’ instructions. DNA amplification, plasmid isolation, E. coli

transformation and electrophoresis were carried out using standard techniques (Sambrook

et al., 2001).

Table 3.1 Bacteria, bacteriophages and plasmids used in this work.

Bacteria, plasmids

and bacteriophages Description

Source and

reference

Bacteria

E. coli JM109 recA1 endA1 gyr96 thi hsdR17 supE44 relA1 Δ(lac-

proAB) [F’ traD36 proAB lacIqZΔM15] Stratagene

Bacillus subtillis

MB24

trpC2 metC3 Laboratory stock

Staphylococcus

aureus RN4220

Restriction deficient derivative of S. aureus

NCTC8325-4 strain capable of receiving foreign

DNA

(Nair et al., 2011)

Streptococcus

pneumoniae Pen6

R6Hex transformant with chromosomal DNA from

penicillin resistant clinical isolate 8249 and selected

for PenR

(Zighelboim et al.,

1980)

Pseudomonas

aeruginosa

ATCC 27853 American type

culture collection

Mycobacterium

smegmatis mc2155

High-transformation-efficiency mutant of M.

smegmatis ATCC 607

(Snapper et al.,

1990)

M. smegmatis

mc2155 (pJV53)

Recombinant strain containing plasmid pJV53 which

expresses recombineering functions

(van Kessel et al.,

2007)

Mycobacterium

tuberculosis

H37Ra

ATCC 25177 American type

culture collection

Mycobacterium

vaccae

SN 901 (IPP)

Institut Pasteur

Production

Mycobacterium

bovis BCG

(Pasteur)

ATCC 35734

American type

culture collection

Plasmids

pQE30 Expression vector; T5 promoter; Ampr Qiagen

pMP302 Ms6 lysB cloned into pQE30 (Gil et al., 2010)

pEGFP-N1 Vector containing the EGFP coding sequence Clonetech

pQE30:egfp egfp fragment cloned into pQE30 This study

pQE30:PGBD-

egfp

Ms6 LysB PGBD fused to egfp inserted into pQE30 This study

Bacteriophages


125

Ms6 Temperate bacteriophage from M. smegmatis. (Portugal et al.,

1989)

Ms6lysB∆PGBD Ms6 derivative mutant lacking the PGBD coding

sequence

This study

Ms6∆lysB 996 bp in-frame deletion of the Ms6 lysB gene (Gigante et al.,

2017)

To construct plasmid pQE30:egfp, a DNA fragment containing the egfp gene was

amplified from pEGFP-N1(Clontech) using primers PrEGFPFw and PrEGFPRv and

inserted into the KpnI and HindIII restrictions sites of vector pQE30 (Qiagen), which

allows the fusion of EGFP with a His6 tag at the N-terminus. Plasmid pQE30:lysBPGBD-

egfp was obtained by amplification of the coding sequence of the first 90 amino acids (the

putative PGBD) of Ms6 LysB with primers PrPGBDFw and PrPGBDRv using Ms6

genomic DNA as template. The PCR product was inserted into the BamHI and KpnI

restriction sites of the previously constructed plasmid pQE30:egfp, allowing the

production of the fusion protein Ms6LysBPGBD-EGFP carrying a N-terminus His6 tag.

All constructs used in this study were verified and validated by nucleotide sequencing.

3.4.4 Protein expression and purification

Expression of the recombinant His6-LysB, His6-LysBPGBD-EGFP or His6-EGFP was

induced from E. coli cells containing plasmids pMP302, pQE30:lysBPGBD-egfp or

pQE30:egfp respectively, with 1 mM IPTG for 4 h. Bacterial cells were then harvested

by centrifugation, washed, resuspended in lysis buffer (50 mM NaH2PO4, 300mM NaCl,

10mM Imidazole, pH 8.0) supplemented with a cocktail of protease inhibitors

(Calbiochem), and disrupted by passage through a French pressure cell press. Cell debris

were removed by centrifugation, and the recombinant proteins were purified in an

AktaPrime Plus® system (GE Healthcare) using a HisTrap FF Crude 1mL Column (GE

Healthcare), according to the manufacturer’s instructions. The eluted protein fractions

were dialysed and stored in PBS at 4ºC, and concentrations were determined by the

Bradford method, using BSA (New England Biolabs) as a standard.


126

3.4.5 Cells binding and fluorescence assay

To test the ability of the recombinant proteins to bind to the bacterial cell surface, M.

smegmatis, M. tuberculosis H37Ra, M. vaccae and M. bovis BCG cells were grown until

OD=±0.8, harvested by centrifugation and washed in SDS 1% in PBS for 1h, with shaking

at room temperature. Cells were then harvested by centrifugation and washed twice with

PBS and incubated with 0.3mg/mL of Ms6LysBPGBD-EGFP or EGFP proteins, for 45

min with shaking at room temperature. Cells were again harvested by centrifugation,

washed trice with PBS, and 3µl were placed on a 1% agar PBS microscope slide. In

parallel assay M. smegmatis cells were pretreated with 0.3mg/mL of Ms6 LysB (instead

of SDS) for 45 min, with shaking at room temperature, followed by the washing steps

describe above. Samples were observed using a Zeiss Axio ObserverZ1 microscope

equipped with a Photometrics CoolSNAP HQ2 camera (Roper Scientific using

Metamorph software, Meta Imaging series 7.5) and analysed using ImageJ software.

3.4.6 Preparation and purification of peptidoglycan

Cell walls of S. aureus, S. pneumoniae, B. subtilis, E. coli and P. aeruginosa were

purified as described before (Filipe et al., 2005), with modifications. Overnight cultures

were subcultured into 1–2 liters of TSB or LB broth (initial OD600 = 0.005) and bacteria

grown until exponential phase (OD600 = 0.5–0.9). Cultures were rapidly cooled in an

ice/ethanol bath and bacteria harvested by centrifugation (13,000 × g, 15 min, 4°C). The

pellet was resuspended in cold ultrapure water and boiled for 30 min with 4% SDS to kill

bacteria and inactivate cell wall-modifying enzymes. To remove the SDS the samples

were centrifuged at 17,000 × g for 10 min. The supernatant was discarded and the pellet

suspended using warm MiliQ H2O, to facilitate the removal of SDS, and again

centrifuged. This washing process was repeated until the supernatant was free of SDS,

accordingly to the Hayashi method (Hayashi, 1975). S. aureus and B. subtilis SDS-free

samples were resuspended in 2 ml of ultrapure water and cell walls disrupted with glass

beads in a homogenizer (FastPrep, Thermo Savant). Fully broken cell walls were

separated from glass beads by filtration (glass filters, pore size: 16–40 μm) and from

unbroken cell walls and other debris by low-speed centrifugation (500 x g, 15 min).

Nucleic acids were degraded after incubation (2 h) at 37°C with DNase (10 μg/ml) and


127

RNase (50 μg/ml) in a buffer containing 50 mM Tris-HCl, pH 7.0, and 20 mM MgSO4.

Proteins were then digested overnight at 37°C with trypsin (100 μg/ml) in the presence

of 10 mM CaCl2. Nuclease and proteases were inactivated by boiling in 1% SDS, and

samples were centrifuged (17,000 x g, 15 min) and washed twice with ultrapure water.

Purified PG from E. coli and P. aeruginosa was lyophilized overnight.

Cell walls from S. aureus and B. subtilis were resuspended and incubated (37°C, 15

min) in 8 M LiCl and then in 100 mM EDTA, pH 7.0, after which they were washed twice

with water. After resuspension in acetone and sonication bath (15 min), cell walls were

washed and resuspended in ultrapure water before undergoing lyophilization.

To obtain purified peptidoglycan from S. aureus and B. subtilis, cell walls (10 mg)

were incubated for 48 h with 4 ml of 46% hydrofluoric acid (HF), under agitation at 4°C.

Samples were washed with 100 mM Tris-HCl, pH 7.0, and centrifuged (17,000 x g, 30

min, 4°C) as many times as necessary to neutralize the pH. The pellet was finally washed

twice with water prior to lyophilization.

To obtain PG from M. smegmatis, the mAGP was first purified accordingly to Bhamidi

et al., (2008). To hydrolyse the mycolic acids, the mAGP was resuspended in 0.5% KOH

in methanol and stirred at 37°C for 4 days. The mixture was centrifuged, and the pellet

was washed twice with methanol and twice with diethyl ether and dried by lyophilisation.

The resulting arabinogalactan-PG was digested with 0.05 N H2SO4 at 37°C for 5 days to

remove the arabinogalactan. The resulting insoluble PG was washed four times by

centrifugation with water and dried.

3.4.7 Peptidoglycan binding assays

90 µg of each purified protein and 25µg of BSA (New England Biolabs) were added

to 200 µg of peptidoglycan resuspended in a final volume of 300 µL, and incubated at

room temperature with shaking, for 45 minutes. The supernatant was recovered by

centrifugation at 350 × g for 10 min, and the pellet was washed and resuspended in 500

µl of PBS and centrifuged at 2000 × g for 5 min, washed again and centrifuged at 5500 ×


128

g for 2 min. Both supernatant and pellet were analysed by SDS-PAGE followed by

Coomassie-blue staining and imaged using a ChemiDoc (Bio-Rad®).

3.4.8 Construction of Ms6lysB∆PGBD

Deletion of 270 bp at the 5’ end of Ms6 lysB gene, corresponding to the putative PGBD

was performed using the BRED technique in M. smegmatis as described previously

(Marinelli et al., 2008). Briefly, a 200-bp dsDNA was constructed from a 100-base

oligonucleotide (lysB∆PGBD) with 50 bases of homology upstream and downstream of

the deleted region, which was extended by PCR using two 75-base flanking primers

PrExt∆lysBFw/PrExtlysB∆PGBDRv, each complementary to 25 bases at each end of the

100-mer. After purification the 200 bp fragment was co-electroporated with 200 ng of

Ms6wt DNA into electrocompetent recombineering M. smegmatis mc2155:pJV53. Cells

were recovered in 1 mL of 7H9 supplemented with 0.5% glucose and 1 mM CaCl2,

incubated at 37°C for 2 h and plated as top agar lawns with M. smegmatis mc2155.

Isolated phage plaques were picked and eluted in 100 μL of phage buffer (10 mM Tris-

HCl, pH 7.5, 10 mM MgSO4, 68.5 mM NaCl,1 mM CaCl2) for 2 h at room temperature

and analysed by PCR with primers PrlysA120Fw/ PrlysBendRv flanking the deleted

region. Mixed primary plaques containing both the mutant and wt DNA were again eluted

as described, and serial dilutions were plated with M. smegmatis mc2155. Individual

secondary plaques were screened by PCR for the presence of pure mutant phages.

3.4.9 One step growth experiments

One-step growth curves assays were adapted from Adams (1959) to mycobacteria and

carried out in exponentials cells. Briefly, 108 M. smegmatis cells were suspended in 1 ml

of a phage suspension at 108 pfu/ml. After 50 min of adsorption at 37º C, non-adsorbed

phages were inactivated with 100 ml of 0.4% H2SO4 for 5 min followed by neutralization

with 100 ml of 0.4% NaOH. The mixture was diluted 1:100 in 7H9 media and aliquots

were taken at intervals of 30 min to quantify the number of phage particles (Catalão et

al., 2010). The obtained results are means of three independent experiments.


129

3.5. References

Adams M. H., (1959) Methods of study of bacterial viruses. Bacteriophages. Adams M.

H. Interscience Publishers, Inc., New York, pp. 443–522.









Bhamidi, S., Scherman, M.S., Rithner, C.D., Prenni, J.E., Chatterjee, D., Khoo, K.H.,

McNeil, M.R., (2008) The identification and location of succinyl residues and the

characterization of the interior arabinan region allow for a model of the complete

primary structure of Mycobacterium tuberculosis mycolyl arabinogalactan. The

Journal of biological chemistry., 283, 12992–13000.

Briers, Y., Volckaert, G., Cornelissen, A., Lagaert, S., Michiels, C.W., Hertveldt, K.,

Lavigne, R., (2007) Muralytic activity and modular structure of the endolysins of

Pseudomonas aeruginosa bacteriophages φKZ and EL. Molecular Microbiology.,

65, 1334–1344.




Chiaradia, L., Lefebvre, C., Parra, J., Marcoux, J., Burlet-Schiltz, O., Etienne, G., Tropis,

M., Daffé, M., (2017) Dissecting the mycobacterial cell envelope and defining the

composition of the native mycomembrane. Scientific reports., 7, 12807.

Filipe, S.R., Tomasz, A., Ligoxygakis, P., (2005) Requirements of peptidoglycan

structure that allow detection by the Drosophila Toll pathway. EMBO reports., 6,

327–333.




130



Gigante, A.M., Hampton, C.M., Dillard, R.S., Gil, F., Catalão, M.J., Moniz-Pereira, J.,

Wright, E.R., Pimentel, M., (2017) The Ms6 Mycolyl-Arabinogalactan Esterase

LysB is Essential for an Efficient Mycobacteriophage-Induced Lysis. Viruses., 9,

343.








Hanych, B., Kȩdzierska, S., Walderich, B., Uznański, B., Taylor, A., (1993) Expression

of the Rz gene and the overlapping Rz1 reading frame present at the right end of

the bacteriophage lambda genome. Gene., 129, 1–8.

Hayashi, K., (1975) A rapid determination of sodium dodecyl sulfate with methylene

blue. Analytical Biochemistry., 67, 503–506.

Henry, M., Coffey, A., O’ Mahony, J., Sleator, R.D., (2011) Comparative modelling of

LysB from the mycobacterial bacteriophage Ardmore. Bioengineered Bugs., 2, 88–

95.

Hildebrand, A., Remmert, M., Biegert, A., Söding, J., (2009) Fast and accurate automatic

structure prediction with HHpred. Proteins: Structure, Function and

Bioinformatics., 77, 128–132.








131

Kedzierska, S., Wawrzynów, A., Taylor, A., (1996) The Rz1 gene product of

bacteriophage lambda is a lipoprotein localized in the outer membrane of

Escherichia coli. Gene., 168, 1–8.

Manganelli, R., Dubnau, E., Tyagi, S., Kramer, F.R., Smith, I., (1999) Differential

expression of 10 sigma factor genes in Mycobacterium tuberculosis. Molecular





3, e3957.

Nair, D., Memmi, G., Hernandez, D., Bard, J., Beaume, M., Gill, S., Francois, P., Cheung,

A.L., (2011) Whole-genome sequencing of Staphylococcus aureus strain RN4220,

a key laboratory strain used in virulence research, identifies mutations that affect

not only virulence factors but also the fitness of the strain. Journal of Bacteriology.










of America., 112, 5497–5502.

Raymond, J.B., Mahapatra, S., Crick, D.C., Pavelka, M.S., (2005) Identification of the

namH gene, encoding the hydroxylase responsible for the N-glycolylation of the

mycobacterial peptidoglycan. The Journal of Biological Chemistry., 280, 326–333.


Mutation., Vol. 18.

Schleifer, K.H., Kandler, O., (1972) Peptidoglycan types of bacterial cell walls and their

taxonomic implications. Bacteriological Reviews., 36, 407–477.


132

Snapper, S.B., Melton, R.E., Mustafa, S., Kieser, T., Jr, W.R.J., (1990) Isolation and

characterization of efficient plasmid transformation mutants of Mycobacterium

smegmatis. Molecular Microbiology., 4, 1911–1919.

Söding, J., Biegert, A., Lupas, A.N., (2005) The HHpred interactive server for protein

homology detection and structure prediction. Nucleic Acids Research., 33.



Biology., 373, 1098–1112.

van Kessel, J.C., Hatfull, G.F., (2007) Recombineering in Mycobacterium tuberculosis.

Nature Methods., 4, 147–152.

Vocadlo, D.J., Davies, G.J., Laine, R., Withers, S.G., (2001) Catalysis by hen egg-white

lysozyme proceeds via a covalent intermediate. Nature.

Vollmer, W., Blanot, D., De Pedro, M.A., (2008)a Peptidoglycan structure and

architecture. FEMS Microbiology Reviews., 32, 149–167.

Vollmer, W., Joris, B., Charlier, P., Foster, S., (2008)b Bacterial peptidoglycan (murein)

hydrolases. FEMS Microbiology Reviews.

Vollmer, W., (2008) Structural variation in the glycan strands of bacterial peptidoglycan.

FEMS Microbiology Reviews., 32, 287–306.

Yamamura, H., Harayama, S., (2007) Method for selective isolation of Mycobacteria

using olive oil emulsified with SDS. Bioscience, biotechnology, and biochemistry.,

71, 1553–1556.



Young, R., (2013) Phage lysis: do we have the hole story yet? Current opinion in


Young, R., (2014) Phage lysis: Three steps, three choices, one outcome. Journal of


Zighelboim, S., Tomasz, A., (1980) Penicillin-binding proteins of multiply antibiotic-

resistant South African strains of Penicillin-Binding Proteins of Multiply

Antibiotic-Resistant South African Strains of Streptococcus pneumoniae, 17, 434–

442.


133




3.6. Supplementary material

Table S3.1 Oligonucleotides used in this work, restriction enzymes sites are underlined.

Name Sequencea Comments

PrPGBDFw GCGGATCCCGCATCGACGGCGAATAC Includes BamHI site to clone in

pQE30

PrPGBDRv GCGGTACCGTCGATGACGGGCCGGGA Includes KpnI site to clone in

pQE30

PrEGFPaFw GCGGTACCGTGAGCAAGGGCGAGGAG

Includes KpnI site to clone egfp in

pQE30 and to generate fusions in

pQE30

Pr EGFP Rv GCAAGCTTCTTGTACAGCTCGTCCATGCC Includes HindIII site to clone in

pQE30

lysB∆PGBD

GGCGGAAAAACCCTCGTGGACGCGGTAGCA

GAACTGTTGGGCCACTGATGACCCGGCCAG

TCCTGTTCACCGTGTGCGGCACCGGCGTGC

CCTGGTGGGT

100bp oligonucleotide to generate

Ms6lysB∆PGBD

PrExt∆lysBFw

AGATCCTGCGGCAACTGCGCGGATACAACC

TCACTGGCTGGCCGCAGCTCGGCGGAAAAA

CCCTCGTGGACGCGG

Extend lysB∆PGBD

PrExtlysB∆PG

BD Rv

CGAGATCCTGCGGCAACTGCGCGGATACAA

CCTCACTGGCTGGCCGCAGCTCGGCGGAAA

AACCCTCGTGGACG

lysB∆PGBD

PrlysA120Fw GCGAGCTCAAGTCACGCATCGCCACCGTC Flanking primer for deletion

screening

PrlysBendRv GCATGGGTACCGCCTCCTATGTGC Flanking primer for deletion

screening


134

Figure S3.1 – Cell binding capacity of EGFP (left column) and Ms6LysBPGBD-EGFP

(right column). (A) Bacillus subtillis MB24; (B) E. coli JM109; (C) Pseudomonas

aeruginosa ATCC 27853; (D) Streptococcus pneumoniae Pen6; (E) Staphylococcus aureus

RN4220. Representative images of each strain are shown in phase-contrast (left) and

fluorescence microscopy (right). Exposure time of 100 ms; scale bar of 10 µm.


135

Figure S3.2 – Negative control of EGFP cell binding capacity. (A) Mycobacterium

smegmatis; (B) Mycobacterium vaccae; (C) Mycobacterium bovis BCG; (D)

Mycobacterium tuberculosis H37Ra. Representative images of each strain are shown in

phase-contrast (left) and fluorescence microscopy (right) . Exposure time of 100 ms; scale

bar of 2 µm indicated for each montage.

CHAPTER 4.

COMPARATIVE MODELLING AND

ENZYMATIC ACTIVITY OF LYSB

MYCOLYLARABINOGALACTAN

ESTERASES FROM MS6, ADJUTOR, TRIXIE

AND U2 MYCOBACTERIOPHAGES


Comparative modelling and enzymatic activity of LysB mycolylarabinogalactan esterases from Ms6, Adjutor, Trixie and U2 mycobacteriophages

Gigante A.M., Gil, F., Moniz-Pereira J., Leandro P., Pimentel M.

Comparative modelling and enzymatic activity of LysB

mycolylarabinogalactan esterases from Ms6, Adjutor, Trixie

and U2 mycobacteriophages

Gigante AM1, Gil F1, Moniz-Pereira J1, Leandro P1 and Pimentel M.1



Running Title: Modelling and activity of LysB proteins

Keywords: mycobacteriophage; mycolil-arabinogalactan esterase, LysB, Ms6, Adjutor, Trixie, U2,

D29


4. Modelling and activity of LysB proteins

141

Acknowledgments

We thank Dr. Graham Hatfull (University of Pittsburgh, USA) for supplying the

mycobacteriophages Adjutor, Trixie and U2. This work was supported by funds provided

by Fundação para a Ciência e Tecnologia (FCT-MCES, Portugal) Grant PTDC/IMI-

MIC/0694/2012 to Madalena Pimentel. Adriano Gigante is a recipient PhD fellowship

(SFRH/BD/87685/2012) from FCT-MCES, Portugal.






143

Abstract

LysB is an esterase enzyme responsible for the hydrolysis of the ester bond between

the peptidoglycan-arabinogalactan complex and mycolic acids, having a role in

mycobacteriophage-induced lysis, by disrupting the mycobacteria outer membrane. A

lysB gene is found in most mycobacteriophage genomes. It is accepted that the LysBs

catalytic core is a triad of serine, aspartic acid and histidine residues, since they belong to

the serine hydrolases family of proteins. Using D29 LysB available structural data, we

successfully modeled Ms6 LysB and discuss the predicted protein structure. Similarly,

we analyzed the LysB proteins of mycobacteriophages Adjutor, Trixie and U2, which

belong to different clusters. All LysBs showed in vitro esterase activity towards short and

medium chain fatty acids esters of p-nitrophenol substrates. In this study we identified

the Ms6 LysB catalytic triad with bioinformatic tools and in vitro enzymatic assays. The

results shown here identify in Ms6 LysB the Ser168-Asp249-His318 as essential for its

esterase activity.


145

4.1. Introduction

Most bacteriophages (or simply phages) synthesize lysis proteins to allow the release

of progeny virions to the extracytoplasmic environment at the end of a replicative cycle.

These proteins are designed to disrupt each layer of the bacterial cell envelope: the holins

are membrane proteins that permeabilize the cytoplasmic membrane defining the timing

of lysis, endolysins are enzymes that cleave one or more bonds of the cell wall

peptidoglycan and spanins produced by phages infecting Gran-negative hosts are

responsible for disruption of the outer membrane (Young & Wang, 2006; São-José et al.,

2007; Berry et al., 2012). Because disruption of the peptidoglycan compromises the

bacterium viability, endolysins have been explored as potential antibacterials against

pathogenic bacteria (Nelson et al., 2012; Schmelcher et al., 2012; Rodríguez-Rubio et

al., 2016).

Mycobacteriophages, the phages that infect mycobacteria, also synthesize holins and

endolysins to compromise the CM and the CW, however disruption of the last layer to

phage release is accomplished by a specific lysis protein, named LysB, in line with the

distinct composition of the mycobacterial OM (outer memebrane)his membrane, also

known as mycomembrane, is a lipid bilayer composed by an outer leaflet rich in lipids

(phospholipids, glycolipids and glycopeptidolipids) non-covalently bound to the cell. The

inner leaflet is formed by long chain fatty acids, the mycolic acids, which are linked via

an ester bond to an arabinogalactan polymer which in turn is covalently linked to

peptidoglycan(McNeil et al., 1990; Brennan et al., 1995; Hoffmann et al., 2008; Zuber et

al., 2008; Jackson, 2014). This outer membrane confers to mycobacteria their

characteristic impermeability to nutrients, resistance to many antibacterial drugs and host

defence mechanisms (Minnikin, 1982; Jarlier et al., 1994).

Mycobacteriophage encoded LysB are proteins with lipolytic activity, which are

members of the serine hydrolase family of proteins. The first description of such activity

for a LysB protein was that of the protein encoded by Ms6 phage, which was shown to

hydrolase the ester bond between arabinogalactan and mycolic acids (Gil et al., 2008,

2010), which result in detachment of the OM from the cell wall. Ms6 LysB has a


146

profound impact on the phage infectious cycle and is considered to be essential for an

efficient host cell lysis (Gigante et al., 2017).

Interestingly, although the mycobacterial OM imposes a restrict access to the

peptidoglycan, activity on mycobacteria cells was reported for the endolysins encoded by

the mycobacteriophages Ms6 and BTCU-1, when exogenously added (Catalão et al.,

2011; Lai et al., 2015). In addition, Lai et al. have reported modifications on the

morphology of M. smegmatis cells exposed to the BCTU-1 LysB protein and a reduced

intracellular survival. However, the mechanisms that ultimately led to these effects are

not yet understood.

A large database of sequenced mycobacteriophage genomes is presently available, and

with that information the number of predicted proteins encoded by those phage genomes

has also increased (http://phagesdb.org). Genes coding for Ms6 LysB homologues are

annotated in most mycobacteriophage genomes, suggesting that LysB is a well conserved

protein. LysB proteins are highly diverse and are organized into phamilies (phams)

according to their amino acid sequence similarity (members of the same pham have a

minimum 32,5 % amino acid identity or a BLAST cutoff below 10-50) (Hatfull, 2014).

Despite the high predicted number of LysB proteins only D29 LysB, besides Ms6 LysB,

have their esterase activity experimental proven (Payne et al., 2009) and to our

knowledge only LysB from D29 had its 3D structure resolved by X-ray crystallography

at 2 Å resolution (PDB ID: 3hc7) (Payne et al., 2009). Using modelling bioinformatics

tools and LysB D29 structural information as the template it is possible to predict other

bacteriophages LysB proteins. This approach was already used to predict Ardmore LysB

structure allowing identification of important characteristics of the protein, such as the

postulated residues involved in catalytic activity (Henry et al., 2011).

In this work the structural data available from LysB D29 was used to modulate and

compare the structure of Ms6, Adjutor, Trixie and U2 LysB proteins. We then performed

in vitro enzymatic assays with low and medium chain fatty acids esters as substrates,

using purified His6-LysB proteins. We confirmed that the new studied LysB proteins

presented esterase activity, although showing different levels of activity. In addition, the

enzymes presented different responses depending on the substrate chain length utilized.


147

Based on the model generated for Ms6 LysB and site-directed mutagenesis we propose

that Ser168-Asp249-His318 form the catalytic triad and show that these residues are

essential for the esterase activity.

4.2. Results

4.2.1 Analysis of LysB sequences alignment from different bacteriophages

shows the GXSXG conserved motif

Putative LysB sequences from several bacteriophages, belonging to different

Clusters/Subclusters, were chosen for an alignment analysis and compared to D29: Ms6

(Cluster F; based on the information available of its lysis cassette), Adjutor (Cluster D),

Trixie (Cluster A2), U2 (Cluster A1), Giles (Cluster Q) and Ardmore (Cluster F1). The

LysB sequences from phage Giles and Ardmore were included in the alignment because

previously data using these proteins have been reported (Payne et al., 2009; Henry et al.,

2011).

From the seven aligned proteins, four (Ms6, Trixie, U2 and Ardmore) have

polypeptide chains of around 330 amino acid residues (322-332), D29 presents only 254

amino acid residues and Giles and Adjutor are expressed as polypeptide chains of 411

and 419 amino acids residues, respectively. Analysis of sequence alignment (Figure 4.1)

indicates that the LysB D29 lacks part of the N-terminal sequence found in Ms6 (90

residues), Adjutor (89 residues), Trixie (69 residues), U2 (78 residues), Giles (101

residues) and Ardmore (90 residues). Sequence alignment also showed a high level of

diversity among the seven LysB proteins as only few amino acids are maintained. From

the aligned sequences Ardmore and Ms6 besides presenting the exact same number of

amino acid residues also showed a remarkably high identity consistent with the fact that

they belong to the same cluster and pham (45104). Even though the LysB proteins have

amino acid sequences with few similarities, they all have in common the conserved motif

GXSXG of lipolytic enzymes (where X is a variable residue), previously reported for

Ms6 LysB, as the pentapeptide motif GYSQG (Figure 4.1, red square) (Gil et al., 2008).


148

Figure 4. 1. Clustal Omega multiple sequence alignment of D29 (UniProtKB - O64205),


149

Ms6 (UniProtKB - Q9FZR9), Adjutor (UniProtKB - B2ZNT7), Trixie (UniProtKB -

G1JV64), U2 (UniProtKB - Q5J5U6), Giles (UniProtKB - A8WA20) and Ardmore

(UniProtKB - D4N7H5) LysB proteins. The red rectangles indicate the regions that

contain the residues assumed to be involved in the catalytic core of LysB proteins. The

three-residue sequence GXP is marked as it is a putative conserved motif in LysB

proteins. The yellow arrows indicate the D29 LysB β-sheets and the red coils indicate the

D29 LysB α-helixes.

It is noteworthy that within the conserved sequence Adjutor LysB motif is different.

Here the sequence GYSQG changed for GYSQK, and the change of a Gly for a Lys in

the last residue of the pentapeptide motif probably will have implications on the enzyme

activity.

Previous studies on D29 LysB showed that the Ser82, Asp166 and His240 residues are

directly involved in the esterase activity presented by the protein (Payne et al., 2009). The

Ser-Asp-His catalytic triad occurs in several protein folds, in fact, it is the most common

catalytic core of α/β hydrolase fold superfamily (Rauwerdink et al., 2015). Previous

alignments of LysB protein sequences revealed that the Ser residue is well conserved in

the Gly-X-Ser-X-Gly pentapeptide motif, which has been previously identified (Gil et

al., 2008; Payne et al., 2009), the Asp residue is highly conserved and there was no

predicted absolutely conserved His residue (Payne et al., 2009; Henry et al., 2011).

However, many LysB proteins have a His residue in the vicinity the D29 LysB His240.

In line with data available for other LysB proteins, alignment of the polypeptide

sequences selected for this study (Figure 4.1) also showed that Ser and Asp of the catalytic

triad are conserved throughout the proteins sequence. Regarding the last amino acid (His)

alignment was only found for Trixie. For Ms6, U2 and Ardmore the His residue shifted

one position to the N-terminus. However, similar to what was found for D29 and Trixie

the preceding residue was always a Pro. The Giles and Ajdutor LysB did not present the

His residue in the postulated position nor in the vicinity, although a Pro was found in the

Giles LysB sequence (PG). According to the localization of the LysB D29 catalytic triad


150

(Ser-Asp-His), the corresponding residues on the studied LysB proteins are expected to

be in positions 168-249-318 (Ms6), 180-272-absent (Adjutor), 149-237-311 (Trixie),

158-238-306 (U2), 194-264-absent (Giles) and 168-249-318.

In addition to the Ser-Asp-His catalytic triad, the GXP motif (where X represents a

variable residue), also reported in the D29 LysB (Payne et al., 2009), has a postulated

involvement in the enzyme mechanism and localizes ~40 residues N-terminally to the

catalytic triad Asp residue. In D29 LysB, the GNP motif (residues 117-119) localizes near

the catalytic triad and defines a turn that allows establishment of H-bonds which may

stabilize the protein structure (Payne et al., 2009). Except Giles LysB (GDP; residues

222-224) all the other sequences aligned with D29 showed the conserved GNP sequence

(Ms6: residues 203-205; Adjutor: residues 227-229; Trixie: residues 184-187; U2:

residues 193-195 and Ardmore: residues 2013-205), probably reflecting the same impact

on overall structure stability.

The analysis performed with Clustal Omega indicates the presence of the motifs

implicated in LysB esterase activity namely the pentapeptide GXSXG sequence

characteristic of lipases and the conserved Ser-Asp-His catalytic triad. However, some

differences in the catalytic activity are expected as a consequence of the unique sequence

of each protein.

4.2.2 Comparative modeling using D29 as the template allowed prediction

of 3D structure of Ms6, Trixie and U2 LysB

Before modeling assessment, the template selection and model quality was estimated

for all the studied proteins. A SWISS-MODEL (Biasini et al., 2014) template search was

performed using the Ms6, Adjutor, Trixie, and U2 LysB protein sequence as query. To

assess the quality of the identified structural templates the GMQE (global model quality

estimation) was taken into consideration with a number between zero and one reflecting

the expected accuracy (higher numbers indicate higher reliability of the model built with

that alignment and template). For Ms6 the best value obtained was 0.58 for D29 LysB

(PDB ID: 3hc7) (Payne et al., 2009). The next higher GMQE obtained was as low as 0.23

for PDB ID: 1XZE (cutinase from Fusarium solani) (Longhi et al., 1996). Also for


151

Adjutor, Trixie and U2 the best values were obtained for D29 LysB with GMQEs of 0.21,

0.73 and 0.59, respectively.

The 3hc7 was retrieved from the SWISS-MODEL Template Library with both BLAST

(Camacho et al., 2009) and HHblits (Remmert et al., 2012) in parallel, as these two

approaches guarantee good alignments at high and low sequence identity levels

respectively. Due to the relatively high sequence similarity score of 0.41 (calculated by

SWISS-MODEL from a normalized BLOSUM62 substitution matrix as the sum of the

substitution scores divided by the number of aligned residue pairs) the BLAST method

was the one employed to match the Ms6 LysB (target) with the 3hc7.1.A (D29 LysB

template).

As 3hc7 demonstrated to be a reliable structural template to predictively evaluate Ms6,

Ajdutor, Trixie and U2 LysB structure divergences, each of the studied LysB sequences

was then modulated with the known 3D structure of D29 LysB (PDB ID: 3hc7) using the

SWISS-MODEL server (Payne et al., 2009).

In Figure 4.2 the overall correspondence between target model (D29 LysB) and

templates is depicted in different colors with blue representing regions with high score

quality, whereas red shows low score regions (based on the QMEAN model quality). For

the modelled Ms6, U2 and Trixie LysB the presented QMEAN4 Z-score, which estimates

the degree of nativeness, was -2.45 (Figure 4.2A, Ms6), -1.71 (Figure 4.2D, U2) and -

1.32 (Figure 4.2C, Trixie) respectively (Figure S4.1). However, for the Adjutor LysB

(Figure 4.2B) a QMEAN4 Z-score of -4.90 was calculated indicating a low quality of the

predicted structure. The obtained values for Ms6, Trixie and U2 LysB are within tolerable

limits indicating that the predicted structures are probably closed to what occurs in nature

(Benkert et al., 2011).

The crystal structure of LysB D29 showed the typical domain organization of the

cutinase family (/ class; / hydrolase fold) with: (i) a central -sheet of five parallel

strands flanked by two parallel -helices (Masaki et al., 2005; Miyakawa et al., 2015);

(ii) the characteristic catalytic triad (Ser-Asp-His) with the residues localized on loops at

the edge of the -sheet core; (iii) a cap/lid domain of -helices extending and covering


152

the -sheet core (Rauwerdink et al., 2015). When looking at the modeled structures they

all presented the -sheet core with a high score (represented in blue, Figure 4.2). The Ms6

structure showed it lowest score regions (predicted local similarity to target <0.6; Figure

S1A) for the residues Asp239-Trp240 (loop before 5), Leu275-Pro279 (G rich loop

before 5) and Met309-Asp320 (N-terminus of 6 and loop connecting to 7). For

Trixie, the predicted local similarity to target lower than 0.6 (Figure S1B) localized on

the loops just before 3 and 5 (Asn139-Asp141 and Met222-Tyr225, respectively. For

U2 the values of predicted local similarity to target lower than 0.6 (Figure S4.1C) were

found for residues Asp100 (N-terminus of 1), Gly207 (loop connecting 4 and 5),

Pro227-Asp228 (loop just before 5), Gly266 (loop rich in G residues, connecting 4

with 5) and Phe298 (N-terminus of 6). As indicated by the QMEAN4 Z score (-4.9),

the Adjutor modeled structure presented predicted local similarity to target lower than 0.6

all over its sequence (Figure S4.1D) and importantly the changes observed for the region

Lys121-Asn147 lead to loss of 2 and changes in 2.


153

Figure 4.2 Structural models of Ms6 LysB (A), Adjutor LysB (B), Trixie LysB (C) and

U2 (D) LysB using D29 LysB (PDB ID: 3hc7) as template. Structures were obtained from

the SWISS-MODEL server. The color of the models is based on the QMEAN score.

Regions in blue have a higher score and are well modelled, and the regions displayed in

red represent the low scored and poorly modelled residues.

In the 3D structure of D29 LysB, Ser82 (catalytic triad) localizes at the C-terminal of the

catalytic core 3 strand in close proximity with Asp166 (loop after 5) and His240

(depicted in Figure 4.3 in red). In Ms6 (Figure 4.3A) Ser168 and Asp249 side chains have

the same orientations than D29 Ser82 and Asp166. However, the His318 residue has a

different positioning. Due to a longer loop between 6 and 7 the His318 side chain is


154

turned inwards and located closer to the Ser168 and Asp249. As expected from the

obtained fitting data, the catalytic triad Ser149-Asp237-His311 in Trixie’s LysB model

(Figure 4.3C) perfectly matches the template provided structure. In U2 LysB modeled

structure (Figure 4.3D) the side chains of the Ser158-Asp238-His306 adopt a positioning

similar to the ones observed for Ms6 with the imidazole chain of His310 facing the protein

core, closer to the catalytic Ser158. The Asp239 residue has a slight discrepancy in 3D

orientation due to the torsion of the loop where it is located. The Ajdutor’s LysB poor

fitting is reflected on the data obtained from super-imposition of the Ser-Asp-His triad

(Figure 4.3B). Not only the side chain of Asp272 is more distant to Ser158, it is

noteworthy that His is absent in the protein sequence.

Figure 4.3 Superimposition of LysB residues of the catalytic triad Ser82, Asp166 and

His240 of D29 with the homologous residues from the modeled structures of Ms6 (A),


155

Adjutor (B), Trixie (C) and U2 (D) LysB. The D29 template residues are shown in red

and the modeled structures in blue. The backbone is represented in light green for the D29

LysB structure and light blue for all the models. Images were obtained using the USF

Chimera package. See text for details of modeling.

Concerning the GNP sequence, it is remarkable that in our models it aligns perfectly

in all protein sequences. Despite previous reports that acknowledge the existence of the

GXP absolutely conserved motif, its true function remains unknown and is yet to be

defined (Payne et al., 2009). Asparagine side chain has a high propensity to establish

hydrogen bonds, both as a donor or acceptor, and in our models the Asn of the GNP motif

is in a loop just after the end of β4 in close proximity to the catalytic Ser (N-terminus of

3). In the α/β hydrolase superfamily, the reaction mechanism involves the oxyanion hole

that via hydrogen bond binds the carbonyl oxygen and stabilizes the oxyanion of the

tetrahedral intermediate (Rauwerdink et al., 2015). We can speculate that the GNP motif

may function as an oxyanion hole, as it is a conserved motif located near the catalytic

triad and able to act as a hydrogen bond donor. In the Ms6 model, similarly to what is

found in the D29 LysB structure, the Asn residue is turned away from the catalytic triad,

being more than 4 Å apart from the expected substrate location. Therefore, this

localization does not suggest a direct intervention in the stabilization of an intermediate

specie during the catalytic reaction. Structural changes after substrate binding might

adjust the GNP residues orientation. However, involvement of this mechanism in LysB

proteins is not yet known.

4.2.3 Relative activity of Ms6, Adjutor, Trixie and U2 LysB

Facing the obtained modelled structures, we then tested the four studied LysB proteins

for their esterase activity. As such, each of the correspondent lysB gene was cloned on

the expression vector pQE30 and the obtained constructs were used to produce the

recombinant LysB proteins.

Under the same expression conditions, soluble proteins were obtained, albeit with

variable yield. Purified proteins (≥ 90% purity grade) were used to determine their

capacity to hydrolyze p-NP esters of fatty acids of different chain lengths (esterase


156

activity). The obtained values are presented in Figure 4 as percentage of the specific

activity of Ms6 LysB, which was considered as 100%.

Figure 4.4 Relative specific activity of LysB proteins towards pNP esters of different

chain lengths. The activity of recombinant LysB proteins from U2 (), Trixie () and

Adjutor () were compared to the values obtained for Ms6 LysB (), which were

considered as 100%. (pNPB) p-nitrophenol butyrate, C4; (pNPL) p-nitrophenol laurate,

C12; (pNPM) p-nitrophenol miristate, C14. The results shown here are means ±SD of

three independent assays.

Under the tested conditions, none of the studied LysB proteins showed a residual activity

higher than Ms6 LysB. U2 LysB presented it highest activity towards the long-chain C14

substrate (66.7± 10.6%) and similar values for the C4 (32.3± 5.3%) and C12 (33.9± 5.4)

esters. The approximate two-fold increase in the residual enzyme activity, when using

pNPM as a substrate, suggest that U2 LysB has higher activity on longer chain esters.

Contrarily, Trixie LysB presented it highest activity towards the C4 substrate. Using

pNPB the obtained relative activity (35.3±8.3%) was approximately three-fold of the


157

values obtained with pNPL (10.3±5.2%) and pNPM (13.7± 9.8%). As such Trixie LysB

has a higher processivity for substrates with shorter carbon chains. The LysB from

Ajdutor phage has the lowest activity of this panel of tested proteins. In fact, with the

utilized spectrophotometric assay it was impossible to measure any released pNP, except

when using pNPM (C14) as substrate, and even though the relative enzymatic activity

was calculated as 5.7±1.0% of Ms6 LysB. The low esterase activity showed by Ajdutor

LysB is in line with the poor model fitting of this protein with the D29 LysB template. If

we take into consideration that the experimental setup was optimized for Ms6 LysB, our

results may indicate that the enzymatic assay for Adjutor LysB may require different

conditions such as pH, temperature or the presence of cofactors.

When comparing the bioinformatics predictions, obtained with LysB proteins modelling,

and the performance results in the activity assays, a correlation between the proteins with

a good model score and higher esterase activity was found. The enzymatic assays

performed with LysB from U2 and Trixie revealed their potential to hydrolase ester

bonds.

4.2.4 Ms6 LysB catalytic triad is formed by Ser168, Asp249 and His318

residues

Among the studied LysB enzymes the recombinant protein from Ms6 presented the

highest activity allowing us to further performed mutagenesis of the residues postulated

to be involved in the mycolylarabinogalactan esterase activity and follow

spectrophotometrically the release of pNP from the C4 substrate (Gil et al., 2008). The

alignment data obtained in this work suggests that in Ms6 LysB the Ser-Asp-His catalytic

triad will localize on residues Ser168, Asp249 and His318. In order to demonstrate the

role of these residues on the Ms6 LysB catalytic activity, site-directed mutagenesis was

performed in order to obtain the Ms6 LysB variants S168A, D249A and H318A.

Among the mutagenized LysB proteins, only the D249A could not be purified as it was

present in the insoluble fraction of E. coli lysate. All the other variants were purified to


158

near homogeneity. Despite several modifications of the expression conditions none could

yield soluble D249A. These results lead us to anticipate that the presence of the Asp

residue in position 249 is essential for Ms6 LysB conformational stability.

The relative enzymatic activity of the designed variant LysB proteins were determined

and compared to the activity presented by the wild type form, which was considered 100%

(Figure 4.5). When using 1 mM of pNPB substrate in the enzymatic reaction, both

variants presented catalytic activities below 1% residual activity (LysB S168A:

0.73±0.66% and LysB H318A: 0.43 ±0.47%). In face of the high SD calculated (probably

reflecting the low sensitivity of the method for the low levels of pNP released) a higher

concentration of substrate was assayed. In fact, when using 5 mM pNPB higher levels of

pNP were released leading to lower SDs. Nevertheless, and despite the slight increase in

the residual activity of the S168A (1±0.079%) and H318A (0.88±0.12) LysB, the

determined values are very low, clearly suggesting that the targeted residues are part of

the catalytic Ser-Asp-His catalytic triad in Ms6 LysB, as they had a strong negative

impact on enzyme activity (S168A and H318A).

Figure 4.5 Relative enzyme activity of wild-type and variant LysB proteins from Ms6,

towards pNPB substrate. The activity of the wild-type (wt) enzyme was considered 100%.


159

The p-nitrophenol butyrate (p-NPB, C4) was used as substrate at 1mM () and 5mM ()

in the final volume of the in vitro assay. The results shown here are the average of three

independent assays ±SD. See text for details

To confirm our hypothesis, additional substitutions were performed on residues

Asp215, Asp306 and His246. These residues are conserved among LysB from D29, Ms6,

Trixie, U2 and Ardmore (Figure 4.1) and can be expected as being part of the catalytic

triad Ser-Asp-His (S168-D215-H246 or S168-D306-H318). The catalytic activities of the

variant LysB proteins, obtained using 1 mM pNPB substrate (Figure 5) were higher than

the ones obtained for the previous variants. In fact the D215A, H246A and D306A

presented residual activities of 20.240.24%, 11.741.82% and 86.423.12%. These

values do not suggest the involvement of Asp215, Asp306 and His246 on the catalytic

triad.

4.2.5 Loss of enzymatic activity of Ala substitutions on Ms6 LysB catalytic

triad residues is corroborated by structural modeling predictions.

The drastic impact on enzyme activity observed for Ala substitutions performed on

Ser168 and His318 of Ms6 LysB can arise either from a direct effect on the chemical

reaction or from a change in the conformation of the catalytic pocket. Here we discuss

the impact of the Ala substitutions on Ms6 LysB regarding the structure of the obtained

model (Figure 4.6).

In the S168A variant many of the hydrogen bonds predicted to be established by

Ser168 are absent in the S168A protein, namely those between Ser168 and Thr101,

Tyr128, Tyr167 and Gly169 (Figure 4.6A and B). This observation anticipates a strong

impact on the local structure, probably destabilizing the loop necessary for substrate

recognition and/or binding.


160

The D249A Ms6 LysB, as mentioned previously, was insoluble and could not be

purified thus preventing its further use on enzyme assays. The predicted hydrogen bonds

lost with the alanine substitution may help to explain such a drastic impact on protein

solubility. The D249A structural model predicts the loss of Asp249 hydrogen bonds with

His246, Val319 and Tyr 321 (Figure 4.6A and C). These residues have an exceptional

negative electrostatic Coulomb’s value (<-10 kcal/(mol·e) at 298 K) and are within loops

located on the protein boundary but not totally exposed at the protein surface. As the

electrostatic interactions between the protein and the solvent are fundamental to define

protein solubility, the model suggests that in the D249A variant His246, Val319 and Tyr

321 may alter the local conformation of the protein exposing these residues at the proteins

surface. As a consequence, the local electrostatic potential at the protein surface will

change thus altering its native interactions with the solvent.

Substitution of His318 by Ala may lead to the loss of the interactions with Tyr167 and

Val319 (Figure 4.6A and D). The His318 in Ms6 LysB model is localized in a loop that

defines the exposed cavity and access to the protein core. The hydrogen bond with Tyr167

possibly stabilizes the His residue in close proximity to Ser168, while the HH bond

His318-Val319 may help to force curvature on the loop defining the cavity limit.

Interestingly the His318 has a predicted HH bond with Tyr251 which is localized in a

very short α-helix near the catalytic Asp249. This bond is absent in the H318A protein

and may be essential for the orientation of the His residue towards the Asp249 and Ser168

catalytic residues.


161

Figure4.6 Structural representation of the modeled wild-type Ms6 LysB (A) and variant

forms S168A (B), D249A (C) and H318A (D). Residues of the catalytic target are shown

in ball and sticks. The interactions established by the target residues are shown in red.

The figures were generated using UCSF Chimera package.

Interestingly for the Asp (215 and 306) and His (246) residues not involved in the

catalytic triad a decrease in the relative enzyme activity was also observed. It is well

known that due to their biological function proteins, and in particular enzymes, are highly

mobile structures that have to adapt to the binding of substrates and to the release of

products. As mentioned before it has been proposed that / hydrolases present also a


162

flexible conformation with two domains. A core domain containing the catalytic

machinery and part of the substrate binding site and a mobile cap/lid domain of -helices

assisting substrate binding (Rauwerdink et al., 2015). As such, it is expected that subtle

changes in local conformation of LysB will lead to changes in enzyme activity. This is

the case of Asp215 which is more than 10 Å away from the catalytic Ser168. According

to the Ms6 structural predicted model, Asp215 has a strong 2.6 Å hydrogen bond with

His246, which in turn establishes a 3.1 Å hydrogen bond with the catalytic Asp249. It is

expected that substitution of Asp215 or His246 by Ala will disrupt the above described

interactions thus destabilizing the local conformation of the protein. In line with these

observations our data showed a higher decrease in the enzymatic activity of H246A when

compared to D215A. Moreover, the Coulomb’s charge of the residues His246, Val319

and Tyr321, located in the vicinity of Asp215 and Asp249 but not totally exposed on the

protein surface, have a predicted exceptional negative electrostatic charge of less than <-

10 kcal/(mol·e). We hypothesize that the decrease in enzyme activity observed for H246A

and D215A may be due to surface electrostatic charge changes, as the result of hydrogen

bonds disruption, which can drastically decrease the enzyme solvation and interaction

with the substrate pNPB.

4.3. Discussion

In the present work we modulated, analyzed and in vitro tested LysB proteins of four

mycobacteriophages from different clusters, for which no structural data is available.

Using the X-ray crystallographic data from the only LysB obtained to date

(mycobacteriophage D29) as the template, predicted structures were obtained for Ms6,

Adjutor, Trixie and U2 LysB proteins and the analysis of each model revealed important

information.

Regardless the high number of predicted LysB homologues, to date only the protein

from Ms6 and D29 have been characterized at an enzymatic level. Data reported for

bacteriophage Giles supports only a biological function as the performed studies on a

constructed Giles mutant phage lacking the lysB gene (GilesΔlysB) demonstrated the

impact of the protein absence on the phage infectious cycle (Payne et al., 2009). For


163

Ardmore LysB the available data regards only to a comparative structural modeling

analysis using LysB 3hc7 structural information as template (Henry et al., 2011). Here

we include, for the first time, data of three more LysB proteins from phages Adjutor,

Trixie and U2 (Hatfull et al., 2010).

The alignment of the selected LysB sequences showed that they present an N-terminal

region which is absent on D29 LysB. It has been reported that LysB from D29 present a

specific enzyme activity towards pNPB (0.72 U.mg-1) higher than the one reported by us

for Ms6 (0.12 U.mg-1). Interestingly the tested proteins from Trixie, U2 and Adjutor

revealed residual activities lower than Ms6 LysB. This data corroborates the hypothesis

that LysB containing the N-terminal region will present catalytic activities lower than the

ones missing this region (as LysB from D29). Alignment also showed other

characteristics of the analyzed sequences such as: a conserved motif GXSXG (GYSQG)

characteristic of lipolytic enzymes and containing the Ser residue of the catalytic triad; a

conserved Asp (position 166 of D29 LysB); a conserved GNP sequence C-terminal to the

GXSSG motif and; a less conserved His residue as part of the catalytic triad (Gil et al.,

2008). From the analyzed LysB, Adjutor protein was the one that presented higher

differences at the level of protein sequence namely a different GYSQG motif (GYSQK)

and the absence of the catalytic triad His. In addition, this protein also presented a low fit

of the modeled structure and a very low catalytic activity. We can postulate that besides

the presence of the Ser and Asp residues involved in the catalytic triad, His will probably

play an important role in catalysis as well on the proper conformation of the GXSXG

motif. The change of a small and flexible residue such as Gly for the larger and positively

charged residue Lys will probably impact the access to the catalytic Ser residue in

Adjutor.

As mentioned before, as proteins classified as / hydrolases, Ms6 LysB activity is

based on the presence of Ser (Ser hydrolases; localized on the GXSXG motif), Asp and

His residues (catalytic triad). Having identified the Ser residue of the GXSXG motif in

position 168, several Asp and His residues could be considered as members of the

catalytic triad namely: Asp215, Asp249, Asp306, His246 and His318. The residual

enzyme activity of variant proteins, where the target residues were substituted by Ala,


164

allowed us to identify Asp166 and His318 as the members of the catalytic triad.

Interestingly the involvement of these residues had been already anticipated by the

alignment data (Figure 4.1) and the modelled structures (Figure 4.3) thus emphasizing the

role and validity of these in silico approaches to identify important residue involved in

protein function.

The overall results presented in this work demonstrate that the mycolyl-

arabinogalactan esterase activity is associated with a catalytic triad formed by Asp Ser

and His residues by this order. The different catalytic activities observed for the LysB

proteins certainly reflect the amino acid sequence diversity observed among them.

The potential application of LysB proteins to disturb the mycobacterial outer

membrane should be investigated. The studies performed with the LysB protein of

mycobacteriophage BTCU-1 showed that when applied exogenously the enzyme was

active against M. smegmatis (Lai et al., 2015). This together with our observation that

Ms6 LysB facilitates the binding of proteins with a PGBD domain to the mycobacterial

cells (Gigante et al., 2018, manuscript in preparation), suggesting that the OM was

sufficiently disturbed to allow labelling of the cells with PGBD containing proteins,

shows that mycobacteriophage-encoded LysB has promising applications. With the

available large collection of mycobacteriophages, LysB activities should be further

investigated. Those with higher catalytic activities are the best candidates to be explored

as potential antibacterials, used alone or in combination with mycobacteriophage

endolysins or conventional antibiotics that target the cell wall, particularly in the case of

antibiotic-resistant bacteria.

4.4. Material and methods

4.4.1 Bacterial strains, bacteriophages and plasmids

Eschericha coli JM109 (recA1 endA1 gyr96 thi hsdR17 supE44 relA1 Δ(lac-proAB)

[F’ traD36 proAB lacIqZΔM15) from Stratagene was used to propagate recombinant

plasmids and for protein expression. The characteristics and sources of the

bacteriophages and plasmids used along the work are indicated in Table 4.1.


165

4.4.2 DNA amplification and cloning

Plasmid extraction and purification, electrophoresis, PCR and transformation by

electroporation was performed using standard techniques (Sambrook et al., 2001).

Restriction enzymes (ThermoScientific®) and T4 DNA ligase (NEB) were used

accordingly to the supplier’s instructions.

In order to construct plasmids pAMG7, pAMG9 and pAMG8, the genes annotated as

lysB were PCR amplified using the respective phage purified DNA from Adjutor, Trixie

and U2 as templates (Table 4.1) and the primers

PrAdjutorBamHIFw/PrAdjutorHindIIIRv, PrTrixieBglIIFw/PrTrixieHindIIIRv and

PrU2BamHIFw/PrU2HindIIIRv (Table S4.1), respectively. The PCR products were

purified with the MinElute PCR Purification kit (Qiagen), digested with BglII and

HindIII, and cloned into the pQE30 BglII and HindIII sites. The resulting plasmids were

transformed by electroporation into E. coli JM109 cells. All recombinants were verified

by DNA sequencing.

Table 4.1 Characteristics and sources of bacteriophages and plasmids used in this work

Name Description Source or reference

Bacteriophage

Ms6 Temperate bacteriophage from M. smegmatis (Portugal et al., 1989)

Adjutor Isolated from M. smegmatis mc2155 (Hatfull et al., 2010)

U2 Isolated from M. smegmatis mc2155 (Hatfull et al., 2010)

Trixie Isolated from M. smegmatis mc2155 (Hatfull et al., 2010)

Plasmids

pQE30 Expression vector; T5 promoter; Ampr Quiagen

pMP302 Ms6 lysB cloned into pQE30 (Gil et al., 2010)

pS168A pMP302 with serine 168 substituted by alanine (Gil, 2011)

pD249A pMP302 with aspartic acid 249 substituted by

alanine This study


166

pH318A pMP302 with histidine 318 substituted by

alanine This study

pAMG7 Adjutor lysB cloned into pQE30 This study

pAMG8 U2 lysB cloned into pQE30 This study

pAMG9 Trixie lysB cloned into pQE30 This study

4.4.3 Site-directed mutagenesis

Site-directed mutagenesis was performed with the QuikChange Site-Directed

Mutagenesis Kit (Agilent Technologies), according to manufacturer’s provided

instructions. Plasmid pMP302 (Table 1) was used as the DNA template to change LysB

Ser, His or Asp target residues to alanine using the mutagenic primers described on Table

S4.1. The correct change was confirmed by sequencing.

4.4.4 Expression and purification of His6-LysB proteins

E. coli JM109 cells were transformed with the desired plasmid (pQE30 derivatives),

grown in LB medium supplemented with ampicillin (100 µg.mL-1) to an OD600 of 0.6.

Expression was induced with 1 mM of IPTG at 37ºC, for 4 h. Bacterial cells were

harvested by centrifugation, washed, resuspended in 50 mM Tris/HCl (pH 7.5)

supplemented with a cocktail of protease inhibitors (Calbiochem), and disrupted by

passage through a French pressure cell press. Cell debris were removed by centrifugation,

and the recombinant His6-LysB present in the supernatant was purified by passage

through a HisTrap FF Crude Ni column (GE Healthcare) in an AKTAPrime

chromatographic system (GE Healthcare). The purification protocol comprised an

imidazole gradient in 50 mM NaH2PO4/Na2HPO4, 300 mM NaCl, pH 8.0 buffer from 20

(washing buffer) to 250 mM (elution buffer) imidazole at a flow rate of 1 mL.min-1. The

protein content of eluted fractions was analyzed by SDS-PAGE, followed by Coomassie

blue staining. The fractions with the highest protein content were dialyzed against PBS at

4ºC for 72h. The purified proteins were quantified by the Bradford method using BSA as

the standard.


167

4.4.5 Catalytic activity assays

The enzymatic assays were performed as described previously by Gil et al., (2008).

Briefly, the assays were performed in a microplate, at room temperature (23ºC), in a final

volume of 200 µL, using 60 µg.mL-1 of recombinant enzyme in 100 mM Tris/HCL (pH

7.5) buffer, supplemented with 0.2% Triton X-100. Substrates with different carbon chain

lengths were used at a final concentration of 1 and 5 mM namely: p-nitrophenyl butyrate

(pNPB, C4), p-nitrophenyl laurate (pNPL, C12) and p-nitrophenyl myristate (pNPM,

C14). The release of pNP from substrates was monitored in a Microplate Reader model

680 (Bio-Rad), over a period of 30 min, at 405 nm. A calibration curve of pNP (0.5– 250

mM) was used to quantify the released pNP. Three independent reactions were performed

for all assays.

4.4.6 In silico analysis

The template selection was performed by SWISS-MODEL against template library

SMTL (Biasini et al., 2014) using both BLAST (Camacho et al., 2009) and HHblits

(Remmert et al., 2012) and the LysB sequences of Ms6 (NCBI accession AAG48319.1),

Adjutor (NCBI accession ACD49625.1), Trixie (NCBI accession AEL17844.1), U2

(NCBI accession AAR89648.1), Giles (NCBI accession ABW88427.1) and Ardmore

(NCBI accession ACY39912.1). Template protein was confirmed and selected as

amongst the Top-ranked by Global Model Quality Estimate (GMQE) (Biasini et al.,

2014).

After template selection SWISS-MODEL online workspace was used to compare and

generate the models of Ms6, Adjutor, Trixie, U2, Giles and Ardmore against

crystallographic data of the template. The query sequence was introduced in FASTA

format, obtained from NCBI database under the respective accession codes. The 3D

protein models were generated based on the target template, engaging the OpenStructure

computational structural biology framework and ProMod3 modelling engine, both used


168

by SWISS-MODEL automatically. The overall quality of the generated model is ranked

by QMEAN scoring function and saved as a pdb file format.

UCSF Chimera was used for structural alignment, amino acid substitutions, hydrogen

bonding and model visualization (Pettersen et al., 2004).

4.5. References

Benkert, P., Biasini, M., Schwede, T., (2011) Toward the estimation of the absolute

quality of individual protein structure models. Bioinformatics., 27, 343–350.



Biasini, M., Bienert, S., Waterhouse, A., Arnold, K., Studer, G., Schmidt, T., Kiefer, F.,

Cassarino, T.G., Bertoni, M., Bordoli, L., Schwede, T., (2014) SWISS-MODEL:

Modelling protein tertiary and quaternary structure using evolutionary information.

Nucleic Acids Research., 42, W252–W258.

Brennan, P.J., Nikaido, H., (1995) The envelope of mycobacteria. Annual review of


Camacho, C., Coulouris, G., Avagyan, V., Ma, N., Papadopoulos, J., Bealer, K., Madden,

T.L., (2009) BLAST plus: architecture and applications. BMC Bioinformatics., 10,

1.

Catalão, M.J., Milho, C., Gil, F., Moniz-Pereira, J., Pimentel, M., (2011) A second



Gigante, A.M., Hampton, C.M., Dillard, R.S., Gil, F., Catalão, M.J., Moniz-Pereira, J.,

Wright, E.R., Pimentel, M., (2017) The Ms6 Mycolyl-Arabinogalactan Esterase

LysB is Essential for an Efficient Mycobacteriophage-Induced Lysis. Viruses., 9,

343.

Gil, F., (2011) Functional analysis of lysB gene from the lysis module of

mycobacteriophage Ms6. PhD Thesis, Univ. Lisboa


169








Hatfull, G.F., Jacobs-Sera, D., Lawrence, J.G., Pope, W.H., Russell, D.A., Ko, C.C.,

Weber, R.J., Patel, M.C., Germane, K.L., Edgar, R.H., Hoyte, N.N., Bowman, C.A.,

Tantoco, A.T., Paladin, E.C., Myers, M.S., Smith, A.L., Grace, M.S., Pham, T.T.,

O’Brien, M.B. et al., (2010) Comparative Genomic Analysis of 60

Mycobacteriophage Genomes: Genome Clustering, Gene Acquisition, and Gene

Size. Journal of Molecular Biology., 397, 119–143.

Hatfull, G.F., (2014) Molecular Genetics of Mycobacteriophages. Microbiology

spectrum., 2, 1–36.

Henry, M., Coffey, A., O’ Mahony, J., Sleator, R.D., (2011) Comparative modelling of

LysB from the mycobacterial bacteriophage Ardmore. Bioengineered Bugs., 2, 88–

95.







Jarlier, V., Nikaido, H., (1994) Mycobacterial cell wall: structure and role in natural

resistance to antibiotics. FEMS microbiology letters.

Lai, M.J., Liu, C.C., Jiang, S.J., Soo, P.C., Tu, M.H., Lee, J.J., Chen, Y.H., Chang, K.C.,

(2015) Antimycobacterial activities of endolysins derived from a

mycobacteriophage, BTCU-1. Molecules., 20, 19277–19290.

Longhi, S., Nicolas, A., Creveld, L., Egmond, M., Verrips, C.T., de Vlieg, J., Martinez,

C., Cambillau, C., (1996) Dynamics ofFusarium solani cutinase investigated


170

through structural comparison among different crystal forms of its variants.

Proteins: Structure, Function, and Genetics., 26, 442–458.

Masaki, K., Kamini, N.R., Ikeda, H., Iefuji, H., (2005) Cutinase-like enzyme from the

yeast Cryptococcus spp. strain S-2 hydrolyzes polylactic acid and other

biodegradable plastics. Applied and Environmental Microbiology., 71, 7548–7550.

McNeil, M., Daffe, M., Brennan, P.J., (1990) Evidence for the nature of the link between

the arabinogalactan and peptidoglycan of mycobacterial cell walls. Journal of

Biological Chemistry., 265, 18200–18206.

Minnikin, D.E., (1982) Lipids: complex lipids, their chemistry, biosynthesis and roles.

In: Ratledge C, Stanford J, eds (ed.), The biology of Mycobacteria. Academic Press,

London.

Miyakawa, T., Mizushima, H., Ohtsuka, J., Oda, M., Kawai, F., Tanokura, M., (2015)

Structural basis for the Ca2+-enhanced thermostability and activity of PET-

degrading cutinase-like enzyme from Saccharomonospora viridis AHK190.

Applied Microbiology and Biotechnology., 99, 4297–4307.

Nelson, D.C., Schmelcher, M., Rodriguez-Rubio, L., Klumpp, J., Pritchard, D.G., Dong,

S., Donovan, D.M., (2012) Endolysins as Antimicrobials. Advances in Virus

Research.



Pettersen, E.F., Goddard, T.D., Huang, C.C., Couch, G.S., Greenblatt, D.M., Meng, E.C.,

Ferrin, T.E., (2004) UCSF Chimera - A visualization system for exploratory

research and analysis. Journal of Computational Chemistry., 25, 1605–1612.




Rauwerdink, A., Kazlauskas, R.J., (2015) How the Same Core Catalytic Machinery

Catalyzes 17 Different Reactions: the Serine-Histidine-Aspartate Catalytic Triad of

α/β-Hydrolase Fold Enzymes. ACS Catalysis., 5, 6153–6176.

Remmert, M., Biegert, A., Hauser, A., Söding, J., (2012) HHblits: lightning-fast iterative

protein sequence searching by HMM-HMM alignment. Nature Methods., 9, 173–

175.


171

Rodríguez-Rubio, L., Gutiérrez, D., Donovan, D.M., Martínez, B., Rodríguez, A., García,

P., (2016) Phage lytic proteins: biotechnological applications beyond clinical

antimicrobials. Critical reviews in biotechnology., 36, 542–552.


Mutation., Vol. 18.

São-José, C., Nascimento, J.G., Parreira, R., Santos, M.A., (2007) Release of progeny

phages from infected cells. Bacteriophage: Genetics and Molecular Biology., pp.

307–334.

Schmelcher, M., Donovan, D.M., Loessner, M.J., (2012) Bacteriophage endolysins as

novel antimicrobials. Future Microbiology.






4.6. Supplementary material

Table S4.1- Oligonucleotides used for cloning and for directed mutagenesis

Oligonucleotides Sequence 5’-3’ Notes

PrS168Ala Fw CACTGGCGGGATATGCGCAAGGCGCG (Gil, 2011)

PrS168Ala Rv CGGGCCTTGCGCATATCCCGCCAGTG (Gil, 2011)

PrD215Ala Fw CACGTTTGGGCTGCTCACGGCGGCTCC (Gil, 2011)

PrD215Ala Rv GGAGCCGCCGTGAGCAGCCCAAACGTG (Gil, 2011)

PrD249Ala Fw GCCCACCAAGGCGCCCTGTACGCCTGC This study

PrD249Ala Rv GCAGGCGTACAGGGCGCCTTGGTGGGC This study

PrD306Ala Fw CCGGGCGATCCTCGCCGCCGGCATGTTC (Gil, 2011)

PrD306Ala Rv GAACATGCCGGCGGCGAGGATCGCCCGG (Gil, 2011)

PrH246A Fw CGCGACTACGCCGCCCAAGGCGACCTG (Gil, 2011)

PrH246A Rv CAGGTCGCCTTGGGCGGCGTAGTCGCG (Gil, 2011)

PrH318A Fw CGCACCGGCCCGGCCGTGGACTACAAC This study


172

PrH318A Rv GGTTGTAGTCCACGGCCGGGCCGGTGCG This study

Pr Trixie BglII Fw CCAGATCTAGCCTCAAGGTCGGTTCC Includes BgllI site to

clone in pQE30

Pr Trixie HindIII Rv GGAAGCTTAGGCTGACGCAGGAAAGC Includes HindIII site

to clone in pQE30

Pr U2 BamHI Fw CAGGATCCCCGCTTAAGCTAGGCGAC Includes BamHI site

to clone in pQE30

Pr U2 HindIII Rv GGAAGCTTAGTCACAGAGCGCAGAGC Includes HindIII site

to clone in pQE30

Pr Adjutor BamHI Fw CCGGATCCGAACTCAAGGTTGGCTCA Includes BamHI site

to clone in pQE30

Pr Adjutor HindIII Rv GTAAGCTTTCGCCGGAAGCCCGCAAC Includes HindIII site

to clone in pQE30


173


174

FigureS4.1. Output of SWISS MODEL using D29 LysB (3hc7) as template to generate the Ms6 LysB (A), Adjutor LysB (B), Trixie LysB (C) and U2 LysB (D) models. See text for modeling details.

CHAPTER 5.

CONCLUDING REMARKS

5. Concluding Remarks

177

5.1. Concluding Remarks

The results presented in this thesis provide additional information that helps to

characterize the role of LysB in mycobacteria phage mediated lysis, and allow the

following conclusion:

• Ms6 LysB is essential for an efficient lysis of Mycobacterium smegmatis host and

by extension for an efficient release of the viral progeny;

• Ms6 LysB N-terminal region shows the ability to bind to mycobacteria cells and

to purified PG, and suffice as functional domain that can be used to construct

fluorescent proteins that target the PG;

• The predicted 3D structure of Ms6 LysB is highly similar to that of D29 LysB,

showing residues S168, D249 and H318 to be the candidates to form the catalytic

triad that is involved in the lipolytic activity.

The results shown here for Ms6 LysB provide new data that can be included into the

state of the art of strategies that phages use to achieve the lysis of their hosts. In this

section, the new findings are integrated and discussed in the light of what was previously

discovered, and how they complement or relate to the established conceptions of

mycobacteriophage Ms6 lytic mechanism. Until recently, the fact that Ms6 LysB is not

necessary for plaque formation, its absence in some mycobacteriophage genomes, and

the lack of known equivalents on other phages led to the wrong assumption that this

protein was not significant for mycobacteriophage mediated lysis.

The construction of a Ms6 mutant lacking the gene coding for LysB, brought new

insights on the impact of LysB in the overall Ms6 mediated host lysis. Absence of LysB

did not prevent host lysis, suggesting that LysB is not essential for lysis. However the one

step experiments performed with Ms6ΔlysB together with the determination of the burst

size showed a defect on the lysis phenotype, with a substantial reduction on the number

of particles released /cell (from 147±27 PFU/cell to 53±14 PFU/cell). Because gene lysB


178

is clustered together with the other lysis genes, it is not expected that the decrease in the

burst size derives from a decrease in the number of synthesized particles but rather would

result from a release deficiency. This was shown to be true by retrieving viable PFU from

the cell pellet following an infection with the mutant phage in a number higher that the

observed for a wt infection. Taking advantage of cryo-EM, lysis of M. smegmatis cells

could be monitored in a time course assay. The cryo-EM data have shown unequivocally

that the Ms6ΔlysB mutant phage cannot accomplish an efficient lysis of the host. At the

time point 180 min part of the viral progeny remains inside incompletely lysed cells in

contrast with Ms6wt where infected cells are completely lysed.

The importance of Ms6 LysB on the phage mediated lysis of M. smegmatis, can be

compared to the spanins of λ on the E. coli lysis. Both LysB or the spanins are required

to overcome the OM, however the mechanism employed is substantially different.

Spanins function results from a conformational change of these proteins which leads to

fusion of the CM and OM while LysB proteins act enzymatically cleaving the link

between the mycobacterial OM and the CW, resulting in a detach of the OM, which

consequently facilitates the release of the viral progeny. (Gil et al., 2010; Cahill et al.,

2017). The ever-increasing number of sequenced mycobacteriophages genomes allowed

to conclude that the lysB gene is simultaneously diverse and commonly found on most

mycobacteriophages genomes. In addition, reinforcing the notion that LysB is important

for efficient host lysis, is the fact that lysB homologues were annotated as Lysin B in

genomes of phages infecting non-mycobacteria, namely gp24 and gp41 from phages

SoilAssassin and Ghobes that infect Gordonae terrae (Pope et al., 2017). Additionally,

LysB-like proteins were also found encoded by phages that infect Tsukamurella species

(Petrovski et al., 2011) and Rhodococcus equi (Summer et al., 2011). These members of

the Corynebacteriales order have a layer of mycolic acids covalently linked to the PG

(Portevin et al., 2004; Petrovski et al., 2011; Dyson et al., 2015), a common characteristic

shared with mycobacteria, often collectively referred to as the mycolate group. It is

reasonable to assume that mycolic acids layer is for phages infecting mycolata bacteria

an additional barrier to overcome, in order to release the viral progeny. Moreover, the

conditions in the natural environment, may favor phages that synthesize LysB or LysB-

like proteins which would improve phage production.


179

The second major conclusion of this work is that the Ms6 LysB N-terminal region

contains a PGBD. The results presented in this thesis clearly demonstrate that the N-

terminus of Ms6 LysB binds to PG, specifically to the A1γ chemotype indicating that the

amino acid sequence of stem peptides and how they cross-link is relevant for the

specificity (Schleifer and Kandler, 1972; Vollmer et al., 2008a). Not surprisingly, a

higher binding affinity to the PG of M. smegmatis was detected by pulldown-assays,

which may be a result of the unusual structural modifications observed in the

mycobacterial PG. These include N-glycolilation of NMurAc, direct diaminopimelic acid

(DAP)-DAP cross-links, and modifications at the free carboxylic acid functions of DAP

and d-Glu. In addition, the C6 of some muramic acid is linked to the galactan domain of

arabinogalactan. How these modifications may contribute to the higher binding affinity

is not known.

Of relevance is that although a PGBD was already predicted for some other LysB

proteins, their ability to bind PG was not so far determined experimentally and in this

work this capacity was demonstrated for the first time for the Ms6 LysB. The high identity

of this region observed among LysB proteins is indicative of a role in PG binding,

however this N-terminus is not present in all mycobacteriophage LysB proteins which

raises the question of what is the importance of this region to host lysis. The one-step

assays using a mutant Ms6lysBΔPGBD show a shorter rise period, this suggests that, in

the assay conditions, the absence of PGBG impacts the lysis progression in a way that

seems to be beneficial for the phage fitness. Theoretically, a shorter rise period allows the

phage to be able to infect more hosts sooner than the other competitor phages, maximizing

the hypothesis to increase the viral population. However, in environmental conditions that

can diverge substantially from the conditions used in the one-step experiment, having a

shorter rise period does not itself provide a competitive advantage at all times.

One can wonder about the importance of PG binding in LysB mAGP esterase function,

and this interrogation remains to be unequivocally replied. Hypothetically: as the N-

terminus of Ms6 LysB binds to the PG, it can be important to localize the protein near the

mycolic acid ester bond with the arabinogalactan; as the 3D structure of Ms6 LysB is not

known, the N-terminus of Ms6 LysB PG binding may be necessary to change the protein


180

conformation in a way that impacts the catalytic domain activity, or the protein as whole;

as it is known Ms6 LysB can interact with the OM from the outside, perhaps the N-

terminus of Ms6 LysB may be necessary to anchor the protein on the cell debris, avoiding

it to diffuse and negatively impact non-infected cells that are potential hosts for the viral

progeny released. All these hypotheses are conceivable, not mutually exclusive and

should be addressed as they can provide us a better understanding of the role of LysB in

mycobacteria phage mediated lysis mechanism.

What is demonstrated in this work, for the first time, is that the N-terminus region of

Ms6 LysB has the ability to bind the PG and, as observed for the fusion PGBD-EGFP, it

suffices to bind PG specially of mycobacteria origin. The PGBD properties, demonstrated

in this work, may be applied in clinical diagnostic or biotechnology (i.e. biosensors). The

detection of mycobacteria by culture methods is time consuming, and the results often

need to be known as quickly as possible. Other authors have published innovative

methods that employ phage encoded proteins to identify Mycobacterium and circumvent

the “time” problem. Arutyunov et al. (2014) have demonstrated that the phage L5 minor

tail protein Gp6 and lysin Gp10 have the ability to bind oligosaccharides, are useful tools

for the rapid capture of mycobacteria, and didn´t recognize E. coli, Salmonella,

Campylobacter or Mycobacterium marinum cells (Arutyunov et al., 2014). Other “phage-

based” diagnostic tools to detect pathogens in food, clinical or veterinary samples have

also been published and subject of review (Kretzer et al., 2007; Schmelcher et al., 2014;

Bai et al., 2016; O’Brien et al., 2018). In the current state however, the PGBD found on

Ms6 LysB needs further testing before an application can be envisioned. It is necessary

to determine in the PG polymer the exact compound that Ms6 LysB binds to, also it is

important to identify if the protein has the ability to recognize and bind other

oligosaccharides or oligopeptides, and the conditions necessary for the binding to occur,

such as pH and temperature.

For most mycobacteriophages a lysB gene was identified in the vicinity of the

endolysin gene. This goes in line with the notion that phages evolve to adapt to specific

hosts by developing specific functions to establish a place in the microbial world and to

guaranty their own survival within the competitive microbial environment. So, after a


181

replicative cycle mycobacteriophages have to face an extremely lipid reach cell envelope,

hence the need to synthesize an enzyme with lipid hydrolase activity that will help phages

to overcome the cell envelope barriers. Enzymes with lipolytic activity share a common

motif G-X-S-X-G where the Ser residue form with an Asp or Glu and His residues a

catalytic triad.

The catalytic region of Ms6 LysB was modeled and, using D29 LysB X-ray

crystallography data as a template, the model of Ms6 LysB was generated. The model

analysis allowed us to conclude that the structure of Ms6 LysB has a high chance to be

similar to the one displayed by D29 LysB. The same procedure was done for the LysB

proteins of phage Adjutor, Trixie and U2. From the model predictions we concluded,

when comparing to the template D29 LysB, that Trixie LysB model has the best fit,

followed by Ms6, U2 and the last is Adjutor LysB. The 3D information of each generated

model was used to infer the position of the serine, aspartic acid and histidine residues that

match the position of D29 LysB residues suggested to be part of the catalytic triad (S82-

D166-H240). We conclude that the triad of catalytic residues is present in LysB of Ms6

(S168-D249-His318), Trixie (S149-D237-His311) and U2 (S158-D239-His310), but the

Adjutor’s model (S180-D272) does not have an histidine in the predicted region. Overall,

the fit of each predicted catalytic residues position of each LysB modelled, followed the

observed score for the whole protein model.

Based on the high sequence diversity among LysB proteins which may account for

different enzymatic activities, LysB proteins of phages Adjutor, Trixie and U2 were

produced and the purified proteins were tested for their enzymatic activity on three

substrates: p-NPB, p-NPL and p-NPM. We concluded that, in the assay conditions used

to determine the relative specific activity: Ms6 LysB has the highest activity against all

substrates; Adjutor LysB has the lowest activity against all substrates; U2 LysB has

increasing activity against substrates with longer carbon chain (p-NPM>p-NPL>p-NPB);

Trixie LysB has increasing activity against substrates with shorter carbon chain (p-

NPB>p-NPL>p-NPM). The main conclusions obtained from the enzymic assays

performed is that, all proteins confirmed the predicted esterase activity and, the

differences in the specific activity observed confirmed how variable the LysB protein


182

family seems to be. Additional assays should be performed regarding the LysB proteins

tested here. One can argue that the optimal conditions required for LysB of Adjutor,

Trixie and U2 and the specific activity against p-NP esters should be determined, and use

only conditions that allow the maximum performance of each enzyme, adjusting factors

such as temperature, pH or de addition of co-factors. However, the assay performed was

the first to compare the specific activity of four different LysB proteins, and the main

objective was achieved: to verify the predicted esterase activity and find new LysB

candidates for further assays, in order to better understand this diverse protein family role,

specifically, on phage mediated host lysis.

The third major conclusion of this work is that the catalytic residues of Ms6 LysB are

S168-D249-His318. The potential catalytic residues in Ms6 LysB were identified based

on two main evidences: the modelling of Ms6 LysB shows the D249 and H318 match

the position of D29 LysB catalytic D166 and H240 residues; the model of Ms6 LysB

S168, D249 and H318 shows these residues are located and oriented towards the protein

center pocket and in close proximity of each other. The comparative enzymic assays of

the mutated LysB proteins showed that the variant protein H318A and S168A have less

than 1% relative activity when compared to the wt protein, and the variant protein D249A

renders the protein insoluble at the expression conditions tested.

Several conserved residues were already proposed to be candidates to be part of the

catalytitc triad (Filipa Gil, PhD Thesis). Site-directed mutagenesis targeting the conserved

residues D215, H246 and D306 showed that the resultant variant proteins have a catalytic

efficiency loss of 87.17% for D215A, 91.49% for H246A and 23.41% for D306A and are

also important for catalysis. We suggest now that the decrease in relative catalytic

efficiency observed for the D215A and H256A variants may be due to the loss of HH

bonds on the ala substituted residues. This can cause a change in the enzyme

conformation, due to electrostatic surface charge decrease as a consequence of exposing

negatively charged residues that were hidden in the protein core, ultimately resulting in a

decrease the enzyme catalytic efficiency.

The state of the art of the lysis mechanism of mycobacteriophages, specifically

regarding Ms6 lysis of M. smegmatis infected cells, is now enhanced by the results


183

showed on this thesis. Ms6 LysB is suggested to be essential for efficient lysis, the

analysis of the protein N-terminus revealed a PGBD, and we improved the understanding

of the catalytic core and overall protein structure. In addition, three different LysB

proteins of three other phages, Adjutor, Trixie and U2 were, for the first time, tested in

this work, adding more candidates to further studies and assays, as until now, to the best

of our knowledge, besides Ms6 LysB, only D29, Giles and Ardmore LysBs were the

subject of published studies.

As a final note, the results showed here are innovative, not only because they suggest

the LysB role as essential for an efficient phage mediated lysis, but also provide insight

of LysB protein for potential future uses, either of its N-terminus PGBD properties or due

to the Ser-Asp-His esterase catalytic domain.

5.2. References

Arutyunov, D., Singh, U., El-Hawiet, A., Dos, H., Seckler, S., Nikjah, S., Joe, M., Bai,

Y., Lowary, T.L., Klassen, J.S., Evoy, S., Szymanski, C.M., (2014)

Mycobacteriophage cell binding proteins for the capture of mycobacteria, 1–9.

Bai, J., Kim, Y.T., Ryu, S., Lee, J.H., (2016) Biocontrol and Rapid Detection of Food-

Borne Pathogens Using Bacteriophages and Endolysins. Frontiers in

Microbiology., 7.

Cahill, J., Rajaure, M., O’Leary, C., Sloan, J., Marrufo, A., Holt, A., Kulkarni, A.,

Hernandez, O., Young, R., (2017) Genetic Analysis of the Lambda Spanins Rz and

Rz1: Identification of Functional Domains. G3 (Bethesda, Md.)., 7, 741–753.

Dyson, Z.A., Tucci, J., Seviour, R.J., Petrovski, S., (2015) Lysis to Kill: Evaluation of the

Lytic Abilities, and Genomics of Nine Bacteriophages Infective for Gordonia spp.

and Their Potential Use in Activated Sludge Foam Biocontrol. (Schuch, R.,

Ed.)PLOS ONE., 10, e0134512.





184

Kretzer, J.W., Lehmann, R., Schmelcher, M., Banz, M., Kim, K.P., Korn, C., Loessner,

M.J., (2007) Use of high-affinity cell wall-binding domains of bacteriophage

endolysins for immobilization and separation of bacterial cells. Applied and

Environmental Microbiology., 73, 1992–2000.

O’Brien, L.M., McAloon, C.G., Stewart, L.D., Strain, S.A.J., Grant, I.R., (2018)

Diagnostic potential of the peptide-mediated magnetic separation (PMS)-phage

assay and PMS-culture to detect Mycobacterium avium subsp. paratuberculosis in

bovine milk samples. Transboundary and Emerging Diseases.

Petrovski, S., Seviour, R.J., Tillett, D., (2011) Genome sequence and characterization of

the Tsukamurella bacteriophage TPA2. Applied and Environmental Microbiology.,

77, 1389–1398.

Pope, W.H., Mavrich, T.N., Garlena, R.A., Guerrero-Bustamante, C.A., Jacobs-Sera, D.,

Montgomery, M.T., Russell, D.A., Warner, M.H., Science Education Alliance-

Phage Hunters Advancing Genomics and Evolutionary Science (SEA-PHAGES),

Hatfull, G.F., (2017) Bacteriophages of Gordonia spp. Display a Spectrum of

Diversity and Genetic Relationships. (Losick, R., Ed.)mBio., 8, e01069-17.

Portevin, D., De Sousa-D’Auria, C., Houssin, C., Grimaldi, C., Chami, M., Daffé, M.,

Guilhot, C., (2004) A polyketide synthase catalyzes the last condensation step of

mycolic acid biosynthesis in mycobacteria and related organisms. Proceedings of

the National Academy of Sciences of the United States of America., 101, 314–319.

Schmelcher, M., Loessner, M.J., (2014) Application of bacteriophages for detection of

foodborne pathogens. Bacteriophage., 4, e28137.






Documents

Mycobacteriophage-mediated lysis: the role of LysB proteins · 2019-03-10 · x Doutora Matilde da Luz dos Santos Duque da Fonseca e Castro, Professora Catedrática da Faculdade de