Dissertations / Theses on the topic 'Endolysins'

There is an increasing incidence of antimicrobial-resistant organisms isolated from food and food products. Coupled with that rising incidence is increased media scrutiny and coverage of outbreaks of foodborne illnesses. Consequently, consumers increasingly demand safer food, and that the antimicrobial measures used be other than antimicrobial drugs. A possible solution is to use bacteriophages, or the purified holin and endolysin proteins that make them lethal and lytic, as antimicrobial food treatments or additives. The bacteriophage Felix O1 is a promising candidate for development as an anti-Salmonella food treatment. This dissertation describes the work done to determine if these proteins could be of value as bioremedial agents. Endolysin treatments of Gram negative bacteria require two agents: the lytic endolysin, and a second agent to permeabilize the outer membrane of the bacterium. The holin protein was proposed as an outer membrane permeabilization agent. Methods used to locate the holin gene included BLAST analysis, analysis of putative Felix O1 proteins for transmembrane domains, and examination of the lysin sequence for an N-terminal signal sequence. Analyses did not reveal a promising candidate. Cloning of rIIA as a potential holin was attempted without success. Results of various analyses are discussed, as are chemical alternatives to the use of purified holin as a permeabilization agent. The endolysin, LysO1, was successfully cloned and characterized. PHYRE analysis predicted that the enzyme structure is composed of α helices arranged into two lobes, with the active site in a cleft between them. The enzyme lysed all tested strains of Salmonella and a tested strain of the foodborne pathogen Escherichia coli. Campylobacter jejuni susceptibility remains ambiguous, and the enzyme had no effect on Listeria monocytogenes or Micrococcus luteus. LysO1 was most active at alkaline pH and low ionic strength; optimal activity was observed in 25 mM buffer at pH 10. If removed from frozen storage, the enzyme was most thermostable at 30 °C. Lytic activity was adversely affected by the presence of the divalent cations calcium, magnesium, and zinc, and by high ionic strength. Considerable time was devoted to development of the activity assay used to further characterize the enzyme, and details of those experiments are provided. Logical extensions of the research project, such as further characterization and testing needed to obtain government approval for widespread use of the treatment, and possible pursuit of treatment based on an enzyme derivative such as an antimicrobial peptide, are discussed.
Ph. D.

The N-terminal hydrophobic domain of the phage P1 endolysin Lyz was found to facilitate the export of Lyz in a sec-dependent fashion, explaining the ability of Lyz to cause lysis of E.coli in the absence of the P1 holin. The N-terminal domain of Lyz is demonstrated to be both necessary and sufficient not only for export to the membrane but also for release into the periplasm of this endolysin. We propose that this unusual N-terminal domain functions as a "signal arrest- release" (SAR) sequence, which first directs the endolysin to the periplasm in membrane-tethered form and then allows it to be released as a soluble active enzyme in the periplasm. To understand why release from the membrane is required for the physiological expression of the lytic activity of Lyz, we examined the role of its seven cysteine residues in the biogenesis of the active endolysin. The inactive, membrane-tethered and the active, soluble forms of Lyz differ in their pattern of intramolecular disulfide bonding. We conclude that the release of Lyz from the membrane leads to an intramolecular thiol-disulfide bond isomerization causing a dramatic conformational change in the Lyz protein. As a result, an active site cleft that is missing in nascent Lyz is generated in the mature form of the endolysin. Examination of the protein sequences of related bacteriophage endolysins suggests that the presence of an SAR sequence is not unique to Lyz. Studies on holin and antiholin indicated that P1 encodes two holins, LydA and LydC. The antiholin LydB inhibits LydA by binding to it directly on the membrane. All above results demonstrate a new paradigm for control of phage lysis, which is, upon depolarization of the membrane by holin function at a programmed time, endolysin is released from the bilayer leading to the immediate lysis of the host.

Clostridium difficile, un bacille à Gram positif anaérobie strict qui forme des spores, est un pathogène opportuniste responsable de simples diarrhées ou de colites pseudomembraneuses qui peuvent provoquer la mort. Le traitement de base réside en l'arrêt des antibiotiques qui ont détruit la flore de l'hôte et provoqué les symptômes, ou en la prescription de vancomycine et/ou de métronidazole. Malheureusement, l'efficacité de ces antibiotiques est variable et le nombre de rechutes est élevé. En outre, de plus en plus de souches deviennent résistantes aux antibiotiques. C'est pour cette raison qu'un besoin d'alternatives thérapeutiques s'est fait ressentir. Une des approches prometteuses est l'utilisation des endolysines, qui sont des enzymes hydrolytiques encodées par les bactériophages et qui se sont déjà révélées être efficaces contre plusieurs bactéries à Gram positif. Dans cette étude, nous avons démontré l'activité lytique d'endolysines encodées par des phages de Clostridium difficile sur des cellules vivantes. Différentes endolysines ont été clonées dans E. coli, exprimées et purifiées, puis leur activité a été vérifiée de plusieurs manières. Certains facteurs biochimiques propres à ces enzymes ont été étudiés, tels que les cofacteurs nécessaires pour une meilleure activité lytique, le pH optimal et le spectre d'efficacité sur différentes souches bactériennes. Finalement, l'étude de ces enzymes comme outil de diagnostic ou de biologie moléculaire est abordée. Les résultats de nos travaux indiquent que les endolysines PlyCD52 et PlyCD38-2 de C. difficile possèdent une faible activité lytique. L'activité des endolysines n'est pas influencée par les cofacteurs Tween 0,5%, Triton 0,1%, MgCl[indice inférieur 2] contrairement à l'EDTA qui inhibe celle-ci. Le pH optimum semble être compris entre 7 et 8,5 et ces enzymes agissent sur différentes souches de C. difficile à l'exception de la souche CD630. Malgré ces résultats encourageants, des travaux supplémentaires seront nécessaires afin de stabiliser les enzymes qui ont une forte tendance à précipiter et d'obtenir une meilleure activité.

In this thesis we focus on the mechanistic understanding of two membrane mediated biological processes: retroviral assembly and endolysin PlyC plasma membrane translocation. Retroviral Gag polyprotein is the structural determinant that assembles in a protein lattice on the hosts plasma membrane to trigger formation of the viral protein/membrane shell.

Dissertação de mestrado em Bioengenharia
Listeria monocytogenes is a Gram-positive opportunistic pathogen that can grow in a wide variety of conditions and is responsible for listeriosis, a potential fatal disease, associated to the ingestion of contaminated food. The concerns about the upsurge of widespread reported cases, combined with emerging antibiotic-resistance amongst pathogenic bacteria, such as L. monocytogenes, demand for the development of novel preservation techniques that ensure the safety of food products. Endolysins, which originate from virulent bacteriophages, are responsible for the hydrolysis of the covalent bonds in peptidoglycan layer of the host cell. These enzyme properties represents a good alternatively approach against Gram-positive foodborne pathogens without altering the organoleptic properties of food products. However, in most of the cases, the activity and stability of naturally occurring enzymes is significantly lower than the biotechnological industry needs. Besides, there is a lack of research advances in lytic activity improvements of endolysins in food storage conditions. The experimental work developed in the scope of this thesis aimed at directing endolysin activity towards refrigeration temperatures against L. monocytogenes through the use of directed evolution strategies – error-prone PCR and cryodrilling. Different attempts were done for the isolation of listerial phages from livestock industries effluents and consequently identification and improvement of lytic activity of its derived endolysins. An in silico analysis of two different lysins – Ply500 and Ply511 – were performed to provide contextualization about their structure and domains. Although both proteins possess modular structure, Ply511 has a central catalytic domain and a not well characterized binding domain which contrasts to Ply500 domain organization. Protein expression in large and micro-scales of wild-type proteins was successfully done and confirmed by performing antibacterial tests against L. monocytogenes 5725. To enhance the activity of endolysins against L. monocytogenes cells, modified endolysins were constructed by amplifying their sequences using error-prone PCR technique and cloning into pQE-30 vector. The cloned vectors were transformed in E. coli JM109 competent cells, however no colonies were obtained. At the same time, using PlyP100 endolysin, a second approach based on the biotic interaction between phage-host at successively temperature was done to promote phage adaptation and consequently enzymatic evolution.
Listeria monocytogenes é um agente patogénico oportunista Gram-positivo, responsável por provocar listeriose, doença potencialmente mortal associada ao consumo de alimentos contaminados. A preocupação inerente à sua capacidade de sobreviver numa grande variedade de condições, o crescente número de surtos da doença e o aumento da resistência a antibióticos obrigam a que novas estratégias de conservação e preservação dos alimentos sejam desenvolvidas. Endolisinas derivadas de fagos são enzimas responsáveis pela lise das células do hospedeiro. Uma vez que não alteram as propriedades organoléticas dos alimentos, o uso destas enzimas representa uma boa alternativa na eliminação de agentes patogénicos Gram-positivos. Contudo, a baixa atividade e estabilidade das enzimas no seu estado natural torna-se incompatível com as necessidades industriais. Este trabalho experimental visou o melhoramento das propriedades líticas das endolisinas a temperaturas de refrigeração recorrendo a técnicas de evolução direta – error-prone PCR e cryodrilling. Numa primeira abordagem foram efectuadas tentativas para o isolamento de fagos de Listeria a partir de efluentes de indústria pecuária com posterior identificação e melhoramento das propriedades líticas das respetivas endolisinas. No entanto, as tentativas não foram bem-sucedidas. Foi efetuada a análise bioinformática das duas diferentes endolisinas – Ply500 and Ply511 – para se obter informações precisas sobre a sua estrutura. Apesar das duas proteínas possuírem uma estrutura modular, Ply511 apresenta um domínio catalítico central com função desconhecida e por conseguinte, pouco caracterizado, relativamente à organização modular de Ply500. A expressão das respetivas endolisinas wild-type em larga e micro escalas foi efetuada com sucesso e confirmada através de testes antibacterianos contra L. monocytogenes 5725. Para melhorar a atividade lítica das respetivas endolisinas, as sequências foram amplificadas por error-prone PCR, clonadas no vetor pQE-30 e transformados em células competentes E. coli JM109. No entanto, não foram obtidas quaisquer colónias. Ao mesmo tempo, usando a endolisina PlyP100, uma nova abordagem baseada no princípio de interação biótica entre fago-hospedeiro, foi efetuada a temperaturas sucessivamente mais baixas de forma a promover a evolução/adaptação do fago e consequentemente da endolisina.

博士
國立中興大學
生物科技學研究所
107
Thermobifida fusca is a moderately thermophilic and cellulolytic actinobacterium. It is of particular interest due to its ability to not only produce a variety of biotechnologically relevant enzymes but also serve as an alternative host for metabolic engineering for the production of valuable chemicals from lignocellulosic agricultural wastes. However, T. fusca lacks a suitable molecular tools and efficient transformation method. The previous studied indicated that bacteriophages played the lead roles in the development of molecular biology and biotechnology. No bacteriophage that infects T. fusca had been reported before this study. Two T. fusca-infecting bacteriophages, named P1312 and P318, were isolated from manure compost and soil in this study. Electron microscopy showed that both of phages have an icosahedral head and a long flexible non-contractile tail, a characteristic of the family Siphoviridae. However, the morphology of P318, including head (40 nm in diameter) and tail (285 nm long and 8 nm wide), is slightly smaller than P1312. P1312 has a double-stranded DNA genome of 60,284 bp with 93 potential ORFs. Nevertheless, P318 has only 48,045 bp long of DNA genome with 52 potential ORFs. The genome arrangement of T. fusca phage was ordered as followed: replication and recombination module, DNA packaging, head morphology, tail morphology, lysis module and lytic/lysogenic module. Surprisingly, modular arrangement of P318 phage was slightly distinguishable to P1312 in the order of DNA packaging module and head morphology. Phylogenetic analysis of large terminase subunit suggests that P1312 is a headful packaging phage containing a chromosome with circularly permuted direct terminal repeats, while P318 was classified under the group of 3’-single stand extended cohesive end. The confirmation of P318 DNA termini was provided by double digestion with BAL-31 and a restriction endoribonuclease to position the ends of chromosome. Besides the difference in DNA packaging mechanisms, the genomes of P318 and P1312 share a low similarity in terms of gene contents, except for endolysin gene and the hypothetical protein (ORF042 of P318 and ORF051 of P1312). The putative endolysin of the bacteriophage P318 consisted of a catalytic domain and a cell‐binding domain. The catalytic domain of P318 endolysin was determined to be an N‐acetylmuramoyl‐L‐alanine amidase and showed the lytic activity against T. fusca and Bacillus circulans. We confirmed that the enzymatic activities of P318 endolysin catalytic domain and cell wall binding domain are independent. The P318 endolysin has also been applied for protoplast preparation of T. fusca. Our experiment is the first report about T. fusca protoplast preparation by bacteriophage endolysin.

Tese de doutoramento em Engenharia Biomédica
The increasing numbers of healthcare-associated infections are alarming and have resulted in higher morbidity and mortality rates, which ultimately are a burden to the public health systems. Remarkably, it was reported that around 65% of these infections were caused by biofilms. Biofilms are microbial communities attached to a surface and surrounded by exopolymeric substances that form a complex structure. Nowadays it is known that the majority of microbes live within a structured biofilm and not as free-floating organisms. Staphylococcus epidermidis is the most predominant species isolated in central line-associated bloodstream infections and central venous catheter-associated bloodstream infections, and second in surgical site infections. Due to the high tolerance of S. epidermidis biofilms to antibiotics, alternative strategies are required to treat these infections. Currently, the use of bacteriophages (phages) is seen as an important strategy to combat pathogenic organisms. Phages are viruses that specifically infect bacteria and are the most abundant entities on our planet, possessing several advantages for therapy, in comparison with antibiotics. Due to the current advances in phage therapy and the almost absence of virulent phages described that infect S. epidermidis, the purpose of this study was the isolation and characterization of new S. epidermidis-specific phages and the assessment of their suitability for controlling S. epidermidis biofilms. Two different phages were isolated from raw effluents of wastewater treatment plants, phiIBB-SEP1 (SEP1) and vB_SepS_SEP9 (SEP9). Phage SEP1 was shown to have typical features of the Myoviridae genus, Twortlikevirus. This phage was shown to be S. epidermidis-specific, being able to kill all the 41 strains tested and in 4 hours, was able to reduce cell viability around 6-log. Phage SEP9 was shown to possess a set of morphological and genomic features that highlights its uniqueness in comparison to all phages described to date. In vitro experiments showed that besides the presence of an integrase, this phage was unable to lysogenize, presenting a virulent behavior. In addition, SEP9 is characterized by a high lytic spectrum being able to infect 28 out of the 38 S. epidermidis strains tested. Furthermore, the presence of a depolymerase suggests its use for therapeutical purposes. As Twortlikevirus are referred as the best candidates for phage therapy, and SEP1 demonstrated good infective properties, this phage was further used in this study. The efficacy of SEP1 in controlling planktonic and biofilm cells of S. epidermidis was assessed. This phage was highly effective in killing planktonic stationary-phase cells. This rare feature makes it a promising biofilm controller due to the heterogeneous physiology of cells within a biofilm. However, this phage was not able to reduce the number of viable cells of 24 h biofilms neither reduce its total biomass. In order to determine if this inefficacy was a consequence of the metabolic activity of the cells within the biofilm and/or a matter of the biofilm matrix, the efficacy of SEP1 was tested against modulated biofilms with distinct metabolic activities (higher or less proportions of dormant cells). The results showed that distinct metabolic activities did not affect phage efficacy. The reduction of activity against planktonic cells upon addition of biofilm matrix, and the efficacy demonstrated by SEP1 against scrapped biofilm cells suggested that the biofilm matrix hinders phage efficacy. Thus, it was hypothesized that S. epidermidis biofilm matrix confers a protective effect against phage predation most probably due to the interaction of phages with PNAG, limiting their access to the cells. Another goal of this thesis was to provide a deeper knowledge about phage endolysins (lysins), which are phage-encoded peptidoglycan hydrolases that are expressed at the end of the phage infection and are promising antimicrobial agents. For that, a database containing all derived lysins from fully sequenced phages were organized and characterized in silico. New catalytic domains and cell binding domains were encountered and described and a new nomenclature was provided, in order to standardize the annotation of endolysins. Finally, the cloning and expression of the first specific endolysins derived from S. epidermidisspecific phages were attempted. Although several approaches to clone these proteins were tested, they were shown to be difficult to express, having solubility or folding problems. In conclusion, the work described in this thesis contributed significantly to increase the knowledge about S. epidermidis phages and endolysins. Furthermore, new insights on phagebiofilm interactions were obtained, which opens new challenges for this field of research.
O numero crescente de infecções associadas à prestação de cuidados de saúde é alarmante e tem resultado em grandes taxas de morbilidade e mortalidade com avultados custos associados. Uma grande percentagem destas infeções, cerca de 65%, são causadas por biofilmes. Biofilmes são definidos como comunidades microbianas aderidas a uma superfície e envoltas numa matriz exopolimérica, formado uma estrutura tridimensional e complexa. Atualmente reconhece-se que a maioria dos microrganismos vive em biofilmes estruturados e não como células livres. Staphylococcus epidermidis é o principal agente etiológico de infecções da corrente sanguínea associadas ao uso de cateteres e em particular ao cateter venoso central, e a segunda espécie mais isolada em infecções no local cirúrgico. Devido ao facto dos biofilmes desta espécie terem elevada tolerância aos antibióticos, é necessário o desenvolvimento de formas alternativas para tratamento destas infecções. Atualmente, o uso de bacteriófagos (fagos) é visto como uma importante estratégia para combater bactérias patogénicas. Os fagos são vírus que infectam especificamente bactérias e são considerados as entidades mais abundantes no nosso planeta. Estes agentes apresentam várias vantagens terapêuticas quando comparados com antibióticos. Devido aos avanços atuais na terapia fágica e à quase ausência de fagos virulentos descritos que infectam S. epidermidis, o objetivo deste estudo foi o isolamento e caracterização de novos fagos específicos para S. epidermidis e à avaliação da sua adequação para o controlo de biofilmes. Dois fagos diferentes foram isolados a partir de efluentes brutos de estações de tratamento de águas residuais – phiIBB-SEP1 (SEP1) e vB_SepS_SEP9 (SEP9). O fago SEP1 mostrou ter características típicas do género Twortlikevirus pertencente à família Myoviridae. Este fago demonstrou ser específico para S. epidermidis, sendo capaz lisar todas as 41 estirpes testadas. Adicionalmente apresentou a capacidade de, em 4 horas, reduzir a viabilidade das células em cerca de 6 log. O outro fago isolado, SEP9, possui características morfológicas e genómicas que destacam a sua singularidade em relação a todos os fagos descritos até ao momento. Experiências in vitro mostraram que este fago tem um comportamento virulento, não obstante apresentar uma integrase no seu genoma. O espectro lítico alargado deste fago (infecta 28 das 38 estirpes de S. epidermidis testadas), bem como a presença de uma depolimerase sugerem a sua utilização para fins terapêuticos. Por outro lado, como os fagos pertencentes ao género Twortlikevirus são referidos como os melhores candidatos para a terapia fágica, o fago SEP1 foi seccionado para ensaios de infecção em biofilmes. A eficácia do fago SEP1 foi avaliada no controlo de células de S. epidermidis quer em fase planctónica quer em biofilmes. Este fago foi altamente eficaz em matar células em fase estacionária. Esta característica rara faz com que seja um fago promissor para controlo de biofilmes, devido à heterogeneidade fisiológica das células dentro de um biofilme. No entanto, este fago foi incapaz de reduzir a biomassa total de biofilmes de 24 horas e de reduzir o seu número de células viáveis. De modo a determinar se esta ineficácia foi uma consequência da atividade metabólica das células dentro do biofilme e/ou resultado da influência da matriz do biofilme, a eficácia do fago SEP1 foi testada em biofilmes modulados com atividades metabólicas diferentes (maior ou menor proporção de células dormentes). Os resultados mostraram que a eficácia fago não é afetada por atividades metabólicas distintas. A redução da atividade detectada contra as células planctónicas, após adição de matriz de biofilme, bem como a eficácia demonstrada contra as células de biofilmes raspados, sugerem que a matriz do biofilme dificulta a eficácia fago. Assim, admitiu-se a hipótese de que a matriz do biofilme de S. epidermidis confere um efeito protetor contra a predação fágica provavelmente devido à interação de fagos com o PNAG, limitando o seu acesso para as células. Outro objetivo deste trabalho foi aprofundar o conhecimento sobre endolisinas (lisinas) fágicas que são hidrolases do peptidoglicano, expressas no final da infecção fágica e que estão descritas como agentes antimicrobianos promissores. Para isso, foi construída uma base de dados contendo todas as lisinas derivadas de fagos totalmente sequenciados através de dados in silico e todas as lisinas foram anotadas. Este trabalho permitiu a identificação de novos domínios catalíticos e domínios de ligação à parede celular. Adicionalmente, com base nos resultados obtidos foi proposta uma nova nomenclatura para uniformizar a anotação de endolisinas. Finalmente, fizeram-se várias tentativas de clonagem e expressão das primeiras lisinas e depolimerases provenientes de fagos específicos para S. epidermidis. Apesar das várias abordagens utilizadas, as proteínas demonstraram ser de difícil expressão, pela baixa solubilidade ou provavelmente pela dificuldade em adquirir a conformação correta. Em conclusão, o trabalho descrito nesta tese contribuiu significativamente para aumentar o conhecimento sobre fagos de S. epidermidis fagos e endolisinas. Além disso, foram obtidos novos conhecimentos sobre interações fago-biofilme, o que poderá abrir novos desafios para esta área de investigação.
FCT Strategic Project PEst-OE/EQB/LA0023/2013 and the Project “BioHealth - Biotechnology and Bioengineering approaches to improve health quality", Ref. NORTE-07-0124-FEDER-000027, co-funded by the Programa Operacional Regional do Norte (ON.2 – O Novo Norte), QREN, FEDER.
The project “Consolidating Research Expertise and Resources on Cellular and Molecular Biotechnology at CEB/IBB”, Ref. FCOMP-01-0124-FEDER-027462 and the FCT grant SFRH/BD/66166/2009.

碩士
慈濟大學
醫學生物技術研究所
99
The improper use of antibiotics nowadays has caused bacteria to resist frequently used antibiotics. These antibiotic-resistant bacteria are called multiple drug-resistant bacteria, particularly Acinetobacter baumannii, has been turning up one after another in a lot of hospitals in Taiwan. When in-patients are infected with multiple drug-resistant Acinetobacter baumannii, MDRAB, it often leads to a situation where no antibiotics can be used to treat the infection. Therefore, it is our aim to develop a new antibacterial agent that can effectively destroy MDRAB. Previous studies show that the endolysin produced after the bacteriophage has infected the bacteria can hydrolyze the peptidoglycan of bacterial cell wall. This suggests that endolysin can perhaps act as an effective antibacterial agent. This study initially used a bioinformatic analysis to find a possible endolysin gene in the A. bumannii phage chromosome. We isolated and characterized a novel phage lysozyme (endolysin) from ?淲B2 and named it LysAB2 To analyze antibacterial activity of LysAB2, the complete LysAB2 and two deletion derivatives were constructed, purified and characterized. Zymographic assays showed that only the intact LysAB2 could lyse the peptidoglycan of A. baumannii and the Staphylococcus aureus cell wall. The bactericidal test results show that LysAB2 can effectively restrain parts of Gram positive and negative bacteria. Under the Scanning Electron Microscope, it was observed that LysAB2 can directly destroy A. baumannii, E. coli, S. aureus. Thermostability assays indicated that LysAB2 was stable at 20~40 ℃. Its optimal pH was 6.0, and it was active from pH 4 to 8. These results indicate that LysAB2 is an effective lysozyme against bacteria, and they suggest that it is a good candidate for a therapeutic/disinfectant agent to control nosocomial infections caused by multiple drug-resistant bacteria.

Thesis for PhD degree in Chemical and Biological Engineeering
Bacteriophages are viruses that specifically infect bacterial hosts to reproduce. At the end of the infection cycle, progeny virions are confronted with a rigid cell wall that impedes their release into the environment. Consequently, bacteriophages encode hydrolytic enzymes, called endolysins, to digest the peptidoglycan and cause bacteriolysis. In contrast to their extensively studied counterparts, active against Gram-positives, endolysins from bacteriophages from a Gram-negative background remain less explored. This knowledge gap is largely due to their limited potential as an antimicrobial, which is related to the presence of an impermeable outer membrane in Gram-negatives that blocks the exogenous endolysin action. The experimental work developed in the scope of this thesis aimed at developing efficient strategies to potentiate the endolysin action against these pathogens. An extensive in silico analysis was performed to provide new insights about endolysins structure and function and bacteriophage-endolysin-host ecology. It was possible to identify and analyze 723 putative endolysins sequences from 5 distinct bacteriophages families, infecting 64 different bacterial genera. These endolysins are tremendously diverse in terms of enzymatic function (24 different enzymatic and 13 binding domains), architecture arrangements (89 different types with either globular or modular design) and length (72 to 578 amino acid residues). Three different novel endolysins (Lys68, ABgp46 and PVP-SE1gp146) were studied in detail. Biochemical characterization of Lys68 (from a Salmonella-infecting bacteriophage) showed that it is highly thermostable, withstanding temperatures up to 100°C, and able to refold to its original conformation upon thermal denaturation. Lys68 was able to lyse a wide panel of Gram-negative bacteria in combination with outer membrane permeabilizers. While the Lys68/EDTA combination could only inactivate Pseudomonas strains, the use of citric or malic acid as permeabilizer broadened and increased its antibacterial effect. Particularly against Salmonella, the combinatory effect of malic or citric acid with Lys68 led to approximately 3 to 5 log reductions after 2 hours, respectively. During an acid-promoted effect, weak acids permeabilized the lipopolysaccharide of most bacteria to Lys68, which retained a relative high activity under these acidic conditions. In case of EDTA, its chelation effect was only observed against Pseudomonas membranes, where ionic interactions are crucial stabilizing forces. The endolysin ABgp46 (from an Acinetobacter-infecting bacteriophage) was shown to naturally inactivate 1 log of certain Acinetobacter strains. Tests in the presence of a number of weak acids (citric, malic, lactic, benzoic and acetic acid) resulted in a powerful antibacterial effect when combined with Abgp46. Higher bactericidal activity was consistently obtained when ABgp46 was combined with citric and malic acid, reducing all planktonic Cronobacter, Klebsiella and E. coli O157 (reductions of 1 to 3 logs) and Pseudomonas, Acinetobacter and Salmonella (reduction of more than 4 logs) species tested. It can be speculated that the major weak acid differences observed are related to their acid dissociated constant, that seems to favor compounds (with lower pKa values) that tend be more ionized. The same combinations did not have significant antibacterial activity when applied against Pseudomonas and Acinetobacter biofilms. To enhance the activity of endolysins against Gram-negative cells, modified endolysins where constructed by fusing PVP-SE1gp146 (from a Salmonella-infecting bacteriophage) with different LPS-destabilizing peptides of polycationic, hydrophobic and amphipathic nature. This strategy resulted in an improvement of the activity of the modified endolysin compared to the native one (1 log reduction on Pseudomonas and Salmonella cells was obtained). The bactericidal activity of all modified variants was increased further in the presence of EDTA. A polycationic nonapeptide was the most efficient tag (maximum reduction of 5 logs). With a different purpose, attempts of increasing the endolysin (in this case Lys68) action against Listeria monocytogenes cells, by inserting species-specific peptidoglycan-binding peptides, did not result in a higher activity. From the moment their genetic identity became known, endolysins have sparked the interest as alternatives for existing antibiotics. Here it was shown that endolysins can be used to kill not only Gram-positive, but also Gram-negative bacterial pathogens. These obtained results underline the great potential of using an endolysin-based strategy for prevention and/or control of Gram-negative pathogens in foodstuff, food processing surfaces, veterinary and medical applications.
Fundação para a Ciência e a Tecnologia grant SFRH/BD/63734/2009 and the projects FCOMP-01-0124-FEDER-019446, FCOMP-01-0124-FEDER-027462 and PEst-OE/EQB/LA0023/2013. Project "Bio-Health - Biotechnology and Bioengineering approaches to improve health quality", ref. NORTE-07-0124-FEDER-000027, co-funded by de Programa Operacional Regional do Norte (ON.2 - O Novo Norte), QREN, FEDER.

碩士
慈濟大學
醫學檢驗生物技術學系醫學生物技術碩士班
102
Tuberculosis is an infectious disease which caused by Mycobacterium tuberculosis. Effective therapies existed are limited by the emergence of multidrug- resistant tuberculosis (MDR-TB) and extensively drug resistant tuberculosis (XDR-TB). Effective therapeutic regimens exist that are limited by the emergence of the inability of antibiotics to kill dormant organisms. Previous studies pointed out that endolysins produced from well assembled phage can hydrolyze the peptidoglycan of cell wall. We have discovered a Mycobacterium smegmatis bacteriophage which genome size of this phage is 46Kb by full genome sequencing and 72 open reading frames were predicted. We selected two suspected endolysin genes named lysA and lysB, cloned into E.coli and expressed, purified the protein to analyze their effect of bactericidal assay. The bactericidal test result showed that LysA and LysB could effectively restrain smegmatis. Under the Scanning Electron Microscope, it was observed that LysA and LysB could directly destroy smegmatis. The cytotoxicity effect of LysA and LysB was determined in HaCaT cell lines by WST-1 assay in vitro, there is no difference was found between negative control. Intracellular bactericidal activity assay showed that treatment of M. smegmatis–infected, RAW 264.7 macrophage, with LysA or LysB, resulted in a significant reduction in the number of viable intracellular bacilli. These results indicate that BTCU-1 and its cloned biologically active lytic modules have antimycobacterial activity, and suggest that they are good candidates for a therapeutic/disinfectant agent to control mycobacterial infections.

Tese de mestrado, Biologia Molecular e Genética, Universidade de Lisboa, Faculdade de Ciências, 2021
Antibiotic resistance is one of the major threats of our time, which has been driving research on antibacterial alternatives. Among these are endolysins, enzymes that digest the peptidoglycan, the major and essential constituent of the bacterial cell wall. Endolysins are produced by bacteriophages, the viruses that infect bacteria, which use them to lyse host cells at the end of the viral replicative cycle for virion progeny release. Those that target Gram-positive bacteria have a typical modular architecture, with one or more catalytic domains responsible for peptidoglycan cleavage in the N-terminal region, and a C-terminal module carrying one or more motifs involved in cell wall binding. Although generally monomeric, two-subunit heteromeric endolysins have already been described. For the recently discovered examples, it was found that the two subunits are produced from a single gene, thanks to the presence of an internal translation start site (ITSS) that allows translation of a smaller reading frame into a C-terminal product of the endolysin. In these cases, the heteromeric endolysins are composed typically by one subunit of the full-length polypeptide, and three copies of the smaller polypeptide. Based on a handful of examples of endolysins with diverse domain architectures that carried putative ITSS, and which are produced by phages that infect clinically or industrially relevant bacteria, we have demonstrated that two different polypeptides are indeed produced from a single gene. These results suggest that this mechanism may be more frequent than anticipated. For one of the studied endolysins, we have also shown that the association of the two polypeptides, with one as a multimer, is necessary for full enzymatic activity. The biochemical analysis also showed that this heteromultimeric endolysin should be composed by one subunit of the full-length polypeptide and between four to six of the smaller polypeptides, indicating a structure never described before.

Tese de doutoramento, Farmácia (Microbiologia), Universidade de Lisboa, Faculdade de Farmácia, 2016
Bacteriophages, or phages, are viruses that strictly infect bacteria. Double-stranded DNA phages use the holin-endolysin system to lyse host cells, thus ensuring the release of the viral progeny and the realization of new infection cycles. The endolysin is an enzyme that degrades the peptidoglycan (PG), the main constituent of the bacteria cell wall (CW). The holin is a transmembrane protein that forms pores in the cytoplasmic membrane (CM), leading to dissipation of the proton motive force (pmf) and consequently to the cell death. The pmf is created by the electrochemical gradient across the CM and is responsible for driving many energy-requiring functions in the cell. According to the way how endolysins reach the CW, these can be classified into two types: i) canonical endolysins (c-endolysins), when access occurs through the holin pores, or ii) exported endolysins (e-endolysins), when transport is performed in a holin-independent pathway. It is considered that once synthesized c-endolysins immediately acquire their active conformation in the cytoplasm, thus having the capacity to effectively degrade the PG if the contact with the CW is allowed. Although exported to the CW by the host cell machinery, all e-endolysins described so far need to be activated by mechanisms that depend on the holin action. Due to their lytic activity, c-endolysins have gained great attention as potential antimicrobial agents for the elimination of pathogenic Gram-positive bacteria, especially in the actual context of increasing resistance to antibiotics. This approach relies on the observation that, at least under certain conditions, c-endolysins are able to efficiently lyse target bacteria when the enzymes are exogenously added in the form of recombinant proteins (enzybiotics). The main objective of the work here presented was to contribute with knowledge for increasing the potential of endolysins as antibacterial agents, namely by developing strategies to improve their production, solubility and lytic performance, and by deepening our understanding of factors and mechanisms that influence their enzymatic activity. This work began with the construction of chimerical endolysins with lytic activity against Staphylococcus aureus (Chapter 2). In addition to obtain enzymes displaying a broad lytic spectrum on clinical strains of this species, we also intended to overcome the problem of low solubility that is commonly observed when overproducing endolysins of S. aureus phages in Escherichia coli. We produced and purified two chimerical proteins (Lys168-87 and Lys170-87) by fusing the same cell wall binding domain (CWBD) of endolysin Lys87, produced by S. aureus phage F87s/06, to the catalytic domain (CD) of endolysins Lys168 or Lys170, produced by Enterococcus faecalis phages F168/08 and F170/08, respectively. This fusion between the CD of highly soluble endolysins and the CWBD of an endolysin with poor solubility, combined with optimized expression conditions, allowed the efficient production of chimerical enzymes in the soluble form. The lytic activity of the chimeras was initially assessed qualitatively by the "spot assay", using bacterial isolates from Portuguese community and hospital settings (n = 100). The endolysins showed high lytic potential, lysing more than 90% of the tested S. aureus isolates, including a high fraction that was methicillin-resistant (MRSA, n = 42). This lytic capacity was also observed in a collection of genetically characterized and typed S. aureus strains, which included representatives of the most relevant MRSA pandemic clones from different parts of the world (n = 30), and representative clones of the dominant methicillin-sensitive S. aureus (MSSA; n = 13). In semi-quantitative assays (lysis curves), Lys168-87 and Lys170-87 were effective in eliminating MRSA strain USA200 suspended in a physiological buffer, being observed a synergistic effect when the chimeras were simultaneously used. Interestingly, unlike the parental endolysins, the chimeras showed a wide lytic spectrum, being also active against Staphylococcus epidermidis, Staphylococcus haemolyticus, Staphylococcus saprophyticus, E. faecalis, Enterococcus faecium and Streptococcus pyogenes. Globally, Lys168-87 presented superior lytic performance than Lys170-87, being this result only inverted when the chimeras were tested on the enterococcal isolates. The results of our research and from many other laboratories support the lytic capacity of c-endolysins and chimeric derivatives when tested in vitro, particularly when the enzymes are added to cells previously suspended in physiological buffers. Nevertheless, we have found that lytic efficacy is often lost or greatly lessened when target bacteria are kept in nutritious media, i.e., in conditions that guarantee maintenance of the pmf and active cell growth. Thus, in a second part of this work we aimed at uncovering factors and mechanisms that underlie this phenomenon. Accordingly, we were led by the fact that in the phage infection context both c- and e-endolysins always act after cells have been killed by the holin. Based on this observation, we decided to study how maintenance or dissipation of the pmf could influence the activity of c-endolysins, either when they reach the CW from the cell inside or when applied externally. The c-endolysin of Bacillus subtilis phage SPP1 (LysSPP1) was chosen as study model, being in a first step transformed into an artificial e-endolysin. For this, the signal sequence (SP) of bacillopeptidase F of B. subtilis (Bpr protein) was fused to the N-terminal end of LysSPP1. The recombinant gene of the artificial e-endolysin SP-LysSPP1 was cloned in a B. subtilis replicative plasmid under the control of an inducible promoter. Surprisingly, the continued production and export of SP-LysSPP1 to the CW, through the secretion system (Sec system) of B. subtilis, produced no obvious effects on the viability and cell growth when compared with a strain that produced the native and not exported LysSPP1. The extracytoplasmic localization and lytic character of the mature form of SP-LysSPP1 was evidenced when cultures were treated with a pmf-dissipating ionophore, gramicidin D. As expected, the addition of the ionophore resulted in immediate growth interruption. However, only the culture producing SP-LysSPP1 showed quick cell lysis after addition of gramicidin D. Similarly, the prior sensitization of B. subtilis with gramicidin D or the maintenance of cells in a buffered medium without energy sources significantly enhanced the lytic activity of the endolysin when added exogenously. It was estimated that the amount of LysSPP1 necessary to lyse cells previously killed by the SPP1 holin action, or by a pmf-dissipating agent, was about 60 times lower than the amount needed to lyse exponentially growing cells. The results followed the same trend when the c-endolysin Lys11, of S. aureus phage 11, was tested in conditions promoting or decreasing the pmf/cell growth. Therefore, the results demonstrate that dissipation of the pmf, which in the context of phage infection occurs by the holin action, can have an activating or potentiating effect on the lytic action of c-endolysins, similarly to what was previously described for e-endolysins. Interestingly, this feature has also been observed with other PG hydrolases, such as autolysins, whose regulation is dependent on the maintenance of the pmf. These results may have implications on the selection and design of endolysins intended for enzybiotic therapy (see Chapter 3). Finally, in a third part of this work we aimed to identify and characterize the holin function of phage SPP1, a necessary condition to perform some of the studies presented in Chapter 3. It was previously proposed that the holin of SPP1 would be encoded by orf 26. Its deduced product (gp26) shares homology and hydrophobic characteristics with the XhlB protein, which was implicated in the holin function of the cryptic phage PBSX of B. subtilis. However, our analysis revealed the upstream orf 24.1 that encodes a holin-like protein analogous to XhlA, also involved in PBSX-mediated lysis. Therefore, we decided to study the role of gp24.1 and gp26 as possible SPP1 holins. Because of its potential toxicity, holins of Gram-positive systems are often studied in heterologous systems, typically E. coli. However, despite some advantages, it is sometimes difficult to extrapolate the results obtained to the native systems. Thus, to understand the contribution of gp24.1 and gp26 to the SPP1holin function, the corresponding orfs were cloned separately and as a transcriptional fusion in a B. subtilis replicative plasmid, under the control of an inducible promoter. The results showed that in the assay conditions the individual production of these proteins did not produce significant impact on B. subtilis cell growth, despite their insertion and accumulation in the CM. Growth cessation and cell death typical of the holin action were only observed after co-production of gp24.1 and gp26, suggesting that in SPP1 the holin function may involve the production of these two proteins. Surprisingly, a constitutive promoter was identified within orf 24.1, which we believe should correspond to the previously described early promoter PE5. The presence of this promoter raises questions regarding lysis regulation in phage SPP1 (see Chapter 4).

Tese de mestrado, Ciências Biofarmacêuticas, Universidade de Lisboa, Faculdade de Farmácia, 2014
Mycobacteriophage Ms6 is a temperate double-stranded DNA (dsDNA) phage that infects the non-pathogenic Mycobacterium smegmatis. Similarly to what happens with all other dsDNA phages studied so far, Ms6 must compromise host cell integrity in order to release its progeny at the end of the lytic cycle. Ms6 lytic operon is organized into five genes. In addition to the endolysin (lysA) and holin-like genes (gp4 and gp5), two accessory lysis genes are found, gp1 and gp3 (lysB), which reflects a novel mechanism of phage-mediated lysis. lysB encodes an enzyme with lipolytic activity whereas gp1 encodes a chaperone-like protein. Gp1 interacts with the N-terminal region of LysA and enables its access to the peptidoglycan layer in a holin-independent manner. However, some aspects concerning Gp1 role in the lytic process are not completely clear. In this work we present data obtained using a recombinant Ms6 carrying gp1 and lysA fused to tag sequences. Subcellular fractionation of M. smegmatis infected cells revealed that Gp1 is present on the cell wall and cell membrane fractions, while LysA seems to be restricted to the cell wall. Despite the association of Gp1 with the cell envelope, translational fusions with the E. coli alkaline phosphatase gene have shown that Gp1 is not endowed with a signal sequence. These results together with the observation that Gp1 is not able to promote the export of the first 60 amino acids of LysA fused to PhoA’ suggest that Gp1 and LysA are exported as a complex. The association between the two proteins may be important to keep LysA inactive until the proper time of lysis. The study of bacteriophages opens new perspectives regarding the treatment of bacterial infections and, in this case, it may also contribute to a better understanding of the diverse mechanisms employed by bacteriophages to lyse their hosts.
Os bacteriófagos, ou fagos, são os vírus que infectam bactérias. Estima-se que os fagos constituem a entidade biológica mais abundante do planeta Terra, desempenhando um papel importante na ecologia e evolução microbianas. Os fagos podem apresentar uma grande variedade de morfologias, no entanto, até à data, a maioria dos fagos descritos apresenta cauda e um genoma de DNA em dupla cadeia (dsDNA). Tal como todos os vírus, os bacteriófagos requerem células hospedeiras para se poderem multiplicar de forma a gerar descendência. De acordo com o seu ciclo de infecção, os bacteriófagos de dsDNA podem ser divididos em virulentos, se realizarem um ciclo lítico, ou temperados, se concretizarem um ciclo lítico ou lisogénico. Durante o ciclo lítico o fago infecta células hospedeiras e multiplica-se, produzindo novas partículas virais no seu interior que vão poder infectar outras células. Para iniciar a infecção o fago deve adsorver à superfície da célula bacteriana através do reconhecimento de receptores específicos presentes no envelope celular e posteriormente injectar o seu genoma na célula hospedeira. Depois da injecção do genoma fágico, este utiliza a maquinaria do hospedeiro de forma a gerar novas partículas virais, que são libertadas durante a lise celular induzida pelo fago. Durante o ciclo lisogénico, o DNA fágico, depois de ser injectado para o interior da célula tal como acontece com os fagos virulentos, é geralmente integrado no genoma do seu hospedeiro, sendo transmitido as células-filhas aquando da divisão celular. Sob determinadas condições o ciclo lítico pode ser induzido e nesse caso o metabolismo do hospedeiro é redireccionado para produzir novas partículas virais. O fim do ciclo lítico culmina com a lise da célula hospedeira para que os viriões recém-sintetizados possam infectar novas células e assim gerar nova descendência fágica. Para que isto aconteça os fagos devem comprometer as estruturas responsáveis pela integridade da célula hospedeira, nomeadamente a parede celular. Os bacteriófagos de dsDNA, como o fago λ, induzem a lise através da síntese de 2 proteínas essenciais: uma endolisina e uma holina. As endolisinas são enzimas com capacidade de hidrolisar o peptidoglicano, enquanto que as holinas são proteínas membranares de pequenas dimensões que conduzem a uma alteração do potencial de membrana e à formação de lesões na membrana citoplasmática, permitindo o acesso da endolisina ao substrato ou a sua activação. As holinas estão descritas como sendo essenciais para determinar o tempo óptimo da lise, de modo a que a libertação de fagos seja produtiva para a sobrevivência do fago. O modelo de lise, holina-dependente, usado pelo fago λ foi por muito tempo considerado universal, no entanto estudos mais recentes realizados com outros fagos têm revelado que o transporte das endolisinas pode ser feito de forma independente das holinas, nomeadamente através dos sistemas de secreção bacterianos. As endolisinas cujo transporte para o meio extracitoplasmático é independente da holina geralmente apresentam uma sequência sinal que permite a translocação da proteína, através da membrana citoplasmática, utilizando o sistema Sec do hospedeiro. A primeira descrição de uma endolisina contendo uma sequência sinal teve origem em estudos com o fago fOg44 de Oenococcus oeni. Neste caso a endolisina (Lys44) é sintetizada com um péptido sinal que é clivado durante o processo de secreção. Mais recentemente têm sido descritas outras endolisinas, nomeadamente de fagos que infectam bactérias Gram-negativas, que apresentam na sua extremidade N-terminal uma região de carácter hidrofóbico, designada por Signal-Arrest-Release (SAR), que permite, da mesma forma, o transporte da endolisina através da membrana celular com o auxílio do sistema Sec. Nestes casos, apesar da holina não apresentar um papel activo no transporte da endolisina, esta apresenta um papel crítico na activação das endolisinas e consequente determinação do tempo de lise ideal. Os fagos que infectam especificamente micobactérias designam-se micobacteriófagos. Este trabalho debruçou-se sobre o bacteriófago Ms6, um micobacteriófago temperado com um genoma de dsDNA que infecta Mycobacterium smegmatis. Tal como todos os outros fagos de dsDNA, o fago Ms6 utiliza a estratégia holina-endolisina para comprometer a integridade celular do seu hospedeiro de forma a libertar a progenia fágica no fim do seu ciclo lítico, no entanto o acesso da endolisina ao peptidoglicano é diferente de todos os modelos descritos até à data. O seu operão lítico está organizado em 5 genes. Para além da endolisina (lysA) e das holinas (gp4 e gp5), existem dois genes adicionais, gp1 e lysB, que são reflexo de um novo mecanismo de lise. lysB codifica uma enzima com actividade lipolítica, enquanto que o gene gp1 codifica uma proteína com características semelhantes às das chaperonas moleculares. O produto do gene gp1, designado Gp1, interage com os primeiros 60 aminoácidos (aa) da região N-terminal da LysA, auxiliando o acesso desta última ao peptidoglicano de forma independente das holinas. No entanto, alguns aspectos relacionados com o papel da Gp1 no processo de lise não são completamente conhecidos. Nomeadamente não se conhece o mecanismo responsável pela manutenção da endolisina num estado inactivo até ao momento da lise determinado pelas holinas. Com este trabalho pretendemos determinar a localização celular da Gp1 e da LysA durante uma infecção de forma a compreender melhor papel da Gp1 no processo de lise. Usando a técnica Bacteriophage Recombineering of Electroporated DNA (BRED) foi construído um micobacteriófago Ms6 recombinante contendo a extremidade 3’ do gene gp1 fundida com um tag c-Myc e a extremidade 3’ do gene lysA com um tag de 6 histidinas (His6). Depois de submeter células de M. smegmatis infectadas com o fago Ms6 gp1-c-Myc lysA-His6 a um protocolo de fraccionamento celular foi possível verificar, por Western-blot, a localização de ambas as proteínas. Apesar da sequência aminoacídica da Gp1 não prever a existência de uma sequência sinal foi possível observar que esta proteína se localiza na parede e membrana celulares. Por outro lado a localização da LysA está restrita à parede celular, o que não é surpreendente uma vez que a endolisina do fago Ms6 possui um domínio de ligação ao peptidoglicano (PGRP) entre os aminoácidos 168 e 312. Para além disso, os resultados mostram que ambas as proteínas começam a ser produzidas antes do tempo de lise e estão ausentes da fracção solúvel. Para verificar a possível existência de um péptido sinal na sequência da Gp1 gerou-se uma estirpe recombinante em que a extremidade 3’ da gp1 está fundida com o gene da fosfatase alcalina sem a sequência sinal (phoA’). Fusões com o gene phoA’ são amplamente usadas para determinar a localização celular de proteínas e a existência de sequências sinal, uma vez que esta enzima só é funcionalmente activa no ambiente oxidativo do periplasma. A ausência de actividade enzimática em meio contendo um substrato cromogénico, bem como em ensaios de quantificação em meio líquido indicam que a Gp1 está desprovida de uma sequência sinal. Por último averiguou-se ainda se a Gp1 tem a capacidade de promover a translocação dos primeiros 60 aa da LysA através da membrana citoplasmática, uma vez que estudos prévios mostram que esta região é essencial para o processo de exportação. Usando a mesma estratégia, construiu-se um plasmídeo em que a sequência que codifica os primeiros 60 aa da LysA está fundida com o gene phoA’ na presença de gp1. Os resultados obtidos indicam que apesar da região N-terminal da LysA ser essencial, esta não é suficiente para promover o transporte da PhoA’ para o espaço periplasmático de M. smegmatis na presença da Gp1. Os resultados obtidos com este estudo parecem sugerir que a Gp1 e a LysA são exportadas em conjunto tal como acontece com outras proteínas secretadas pelas micobactérias. Para além disso, a formação do complexo Gp1-LysA parece ser importante para o processo de translocação. De acordo com estas observações colocamos a hipótese de que a Gp1 poderá estar envolvida na manutenção da LysA num estado inactivo até ao momento de lise, uma vez que ambas as proteínas são exportadas enquanto estão a ser sintetizadas, no entanto são necessários estudos adicionais para confirmar esta hipótese. O estudo dos mecanismos de lise usados pelos bacteriófagos abre novas perspectivas no que diz respeito ao tratamento de infecções bacterianas. Para além disso, o estudo da cassete de lise do micobacteriófago Ms6 contribui para uma melhor compreensão dos diversos mecanismos usados pelos bacteriófagos para lisar os seus hospedeiros e lança novas questões relativamente aos mecanismos de secreção usados pelas micobactérias.

SAR endolysins are a recently discovered class of muralytic enzymes that are regulated by dynamic membrane topology. They are synthesized as enzymatically inactive integral membrane proteins during the phage infection cycle and then are activated by conformational remodeling upon release from the membrane. This topological duality depends on N-terminal SAR (Signal-Anchor-Release) domains, which are enriched in weakly hydrophobic residues and require the proton motive force to be maintained in the bilayer. The first SAR endolysin to be characterized was P1 Lyz, of phage P1. Its activation requires a disulfide bond isomerization involving its catalytic Cys initiated by a free cysteine thiol from the newly-liberated SAR domain. A second mode of disulfide bond regulation, as typified by Lyz103 of the Erwinia Amylovora phage ERA103, has been demonstrated. In its membrane bound form, Lyz103 is inactivated by a disulfide that is formed between cysteine residues flanking a catalytic glutamate. A second class of SAR endolysins, typified by R21, the lysozyme of the lambdoid phage 21, does not require disulfide bond isomerization for activation. Rather, these proteins are dependent on the release of the SAR domain for proper folding of the catalytic cleft. Bioinformatic analysis indicates that the regulatory theme of R21 is common in the SAR endolysins of dsDNA phages. Furthermore, bioinformatic study of endolysins of dsDNA phage of Gram-negative hosts revealed two new classes of SAR endolysins that are not homologous to T4 gpe, as all SAR endolysins were once thought to be. SAR endolysins were found in nearly 25% of sequenced dsDNA phages of Gram- negative hosts including 933W, which is involved in the release of Shiga toxin from EHEC strain EDL933. An inhibitor study against the SAR endolysin of 933W, R933W, was performed using a custom compound library in a high through-put, in vivo lysis assay. Of nearly 8,000 compounds screened, one compound, designated 67-J8, inhibited lysis but not growth. In vivo and in vitro experiments show that the compound has no effect on R933W activity, accumulation, or secretion. In vivo experiments suggest that 67-J8 increases the proton motive force, thereby presumably retaining the SAR domain in the membrane.

碩士
臺灣大學
微生物學研究所
95
Comparison of the genome sequences of Klebsiella pneumoniae NTUH-K2044 strain with those of MGH 78578 strain, we found that there is a different region (about 25 kb) in the 2089556 to 2114933 sites of the genome of NTUH-K2044 strain. Analysis of the different region by using bioinformatics tools, National center for biotechnology information Basic Local Alignment Search Tool (NCBI BLAST), we found that there is an open reading frame being predicted as an endolysin gene. Endolysin is the enzyme encoded by bacteriophage and directly targets bonds in the peptidoglycan layer of the bacterial cell wall. The result of this activity is degradation of the rigid murein layer and release of newly assembled virions at the terminal stage of the phage reproduction cycle. According to previous reports, the capability of endolysin to digest the cell wall when applying exogenously to bacterial cells has enabled its use as alternative antibacterials. The putative endolysin gene was clone to Escherichia coli, and protein was expressed and purificated. And then we applied the purified protein exogenously to bacterial cells, and their survival rate was observed. The results showed that 50 μg purified protein can decrease the survival rate of Escherichia coli to 7.22% in 15 minutes. On the other hand, the protein also showed selective bactericidal activity. Among the tested strains, Escherichia coli is the most sensitive to the protein, Klebsiella pneumoniae is the second, and Bacillus subtilis is the least sensitive to the putative endolysin. In order to find out if peptidoglycan is the substrate of the protein, we performed peptidoglycan degrading assay. The results showed that the protein can not degrade peptidoglycan of Bacillus subtilis. It may be other components of bacteria that the protein targets, and results in the bactericidal effect.

The growing threat of drug- resistant Staphylococcus aureus (S. aureus) infections mandates the need to develop novel, effective and alternative antibacterial therapeutics. Despite infection prevention and control measures, methicillin resistant S. aureus (MRSA)-associated deaths reached 11,285 in 2011 in the USA (CDC, 2013). To counteract the threat of drug resistant S. aureus, we sought to construct and characterize a novel therapeutic based on the display of lytic antibacterial enzymes, termed endolysins. These endolysins were displayed on the surface of a specific bacterial virus, bacteriophage (phage), to generate lytic antibacterial nanoparticles. Endolysins are encoded individually by a variety of double-stranded DNA phage and act to direct host lysis and escape. These lytic enzymes confer a high degree of host specificity that could potentially substitute for, or be combined with, antibiotics in the treatment of gram-positive drug resistant bacterial infections such as MRSA. In this study, modular domains of the phage-encoded endolysin K enzyme, specific to S. aureus, were displayed on the capsid surface of phage lambda () via fusion with the λ major head (capsid) protein, gpD. The constructs of displayed endolysins were prepared in various combinations to maximize the functional display of gpD::X fusions on the phage. Phage lysates were generated, collected and purified and lysis was investigated by adding to fresh lawns of MRSA, vancomycin resistant S. aureus (VRSA) and bovine S. aureus. Phage preparations did not readily confer cell lysis, likely due to poor incorporation of the fusions onto the functional phage capsid. We purified the fusion proteins (gpD::X) and tested them for their lytic activity. We noted that the activity of the gpD::LysK protein was not impaired by the fusion and demonstrated lysis on live and dead (autoclaved) bovine S. aureus. In contrast to gpD::LysK, the gpD::CHAP protein fusion, expressing only the CHAP catalytic domain of endolysin K showed variable results in the lysis assays that we performed. In the zymogram assay, gpD::CHAP did not elicit any observable lysis on live bovine S. aureus cells, but did effectively lyse dead cells of the same S. aureus species; however, it was highly lytic in the inhibition assay on bovine S. aureus. The CHAP::gpD protein fusion, which is the CHAP domain fused to the N terminus of gpD only showed its ability to inhibit bovine S. aureus growth on the inhibition assay. The fusion of endolysin K or its CHAP domain to gpD protein does not seem to interfere with lytic activity, but may result in recalcitrant gpD fusions that compromise the ability to efficiently decorate the phage capsid. Suggestions for improved fusion capsid integration are discussed.

Dissertation for Master Degree in Biotechnology Supervisor
Acinetobacter species have become a major cause of nosocomial infections worldwide due to its inherent and adaptive resistance towards the majority of antibiotics, survival in hostile environments and virulence traits. The increased mortality rates associated with these hard-to-treat bacteria urgently demands the development of new therapeutic options. In the present work, bacteriophage endolysins and depolymerases were explored to tackle Acinetobacter infections. Five novel enzymes were identified, heterologously expressed in Escherichia coli and characterized to gain insights into their enzymatic and structural features, as well tested in vitro (Acinetobacter cells and human cell line A549) and in vivo (Galleria mellonella) to assess their therapeutic potential. Remarkably, endolysins exhibited intrinsic ability to kill Acinetobacter cells, with optimal activity at pH 5-7 and low ionic strength (20 mM), reducing approximately 3.5 log to > 5 log (below the detection limit) within 2 h. Interestingly, no antibacterial activity of endolysins could be observed in biomatrices such as culture medium. All depolymerases exhibited activity on extracellular polysaccharides in all ranges of pH values (pH 5 to 9) and ionic strengths (0 to 500 mM) and retained at least 50% of the activity at 70 ºC. Consistently, circular dichroism analysis showed moderate and high thermostability of endolysins (rich in α-helices) and depolymerases (rich in β-sheets), with melting temperatures around 56 ºC and > 68 ºC, respectively. Combinations of endolysin and depolymerase reduced 24 h-old bacterial biofilms in approximately 1 log. Further work conducted on depolymerases demonstrated that these enzymes degrade the bacterial capsular polysaccharides (observations confirmed by Atomic Force Microscope). The removal of the capsule via depolymerase activity reduced the virulence of Acinetobacter to human epithelial cells for 2 h. Complementary studies performed in vivo showed that depolymerase was able to rescue Galleria mellonella larvae from Acinetobacter baumannii infection in a time- and dose-dependent manner. Maximal difference survival between larvae injected with depolymerase-treated bacteria and non-treated bacteria was observed after 72 h with 5 μM of depolymerase (88% vs 10%; P < 0.01). In summary, the enzymes studied in this thesis revealed to be effective candidates to control problematic Acinetobacter infections, either as bacteriolytic or anti-virulence agents.
As espécies de Acinetobacter tornaram-se uma das principais causas de infeções nosocomiais em todo o mundo, devido à sua resistência inerente e adaptativa à grande maioria dos antibióticos, sobrevivência em ambientes hostis e fatores de virulência. O aumento da taxa de mortalidade associada com estas bactérias de difícil tratamento exige, urgentemente, o desenvolvimento de novas opções terapêuticas. No presente trabalho, endolisinas e depolimerases de bacteriófago foram exploradas para combater infeções de Acinetobacter. Cinco novas enzimas foram identificadas, expressas de forma heteróloga em Escherichia coli e caracterizadas para obter informações sobre as suas características enzimáticas e estruturais, e também, testadas in vitro (células de Acinetobacter e linha celular humana A549) e in vivo (Galleria Mellonella) para avaliar o seu potencial terapêutico. Notavelmente, as endolisinas exibiram capacidade intrínseca para matar células de Acinetobacter, com atividade ótima no pH 5-7 e a baixa força iónica (20 mM), reduzindo aproximadamente 3.5 log até > 5 log (abaixo do limite de deteção) em 2 h. Curiosamente, nenhuma atividade antibacteriana das endolisinas foi observada em biomatrizes, como meio de cultura. Todas as depolimerases exibiram atividade contra os polissacarídeos extracelulares em todas as gamas de valores de pH (pH 5 a 9) e força iónica (0 a 500 mM) e conservaram pelo menos 50% da atividade a 70 ºC. De forma consistente, os estudos em dicroísmo circular mostraram moderada e alta thermoestabilidade das endolisinas (ricas em α-helices) and depolimerases (ricas em β-sheets), com temperaturas de desnaturação de aproximadamente 56 ºC and > 68 ºC, respetivamente. Combinações de endolisina e depolimerase reduziram biofilmes bacterianos de 24 h em aproximadamente 1 log. Estudos adicionais realizados com as depolimerases demonstraram que estas enzimas degradam a cápsula polissacarídica bacteriana (observações confirmadas por microscopia de força atómica). A remoção da cápsula pela atividade da depolimerase reduziu a virulência de Acinetobacter em células epiteliais humanas por 2 h. Estudos complementares realizados in vivo revelaram que a depolimerase foi capaz de resgatar larvas de Galleria Mellonella da infeção de Acinetobacter baumannii, de uma forma dependente do tempo e da dose. A maior diferença de sobrevivência entre larvas injetadas com bactéria tratada com depolimerase e bactéria não tratada foi observada após 72 h com 5 μM de depolimerase (88% vs 10%; P < 0.01). Em suma, as enzimas estudadas nesta tese revelaram ser candidatas eficientes para o controlo de infeções problemáticas de Acinetobacter, quer como agentes antibacteriano ou anti-virulência.
This study was supported by the Portuguese Foundation for Science and Technology (FCT) under the scope of the strategic funding of UID/BIO/04469/2013 unit and COMPETE 2020 (POCI-01-0145-FEDER-006684) and BioTecNorte operation (NORTE-01-0145-FEDER-000004) funded by the European Regional Development Fund under the scope of Norte2020 - Programa Operacional Regional do Norte and the Project PTDC/BBB-BSS/6471/2014 (FEDER: POCI-01-0145-FEDER-016643).