Fauna Acompanhante: Um Universo Químico a Ser Explorado

UNIVERSIDADE ESTADUAL PAULISTA “Júlio de Mesquita Filho” Instituto de Química - Campus Araraquara

Marcelo Marucci Pereira Tangerina

Tese de Doutorado apresentada como parte dos requisitos para a obtenção do título de Doutor em Química.

Orientador: Prof. Dr. Wagner Vilegas Co-Orientador: Prof. Dr. Wagner C. Valenti

Araraquara 2016

FICHA CATALOGRÁFICA

Tangerina, Marcelo Marucci Pereira T164b By-catch: a chemical universe to be explored = Fauna acompanhante um universo químico a ser explorado / Marcelo Marucci Pereira Tangerina. – Araraquara: [s.n.], 2016 122 p.: il.

Tese (doutorado) – Universidade Estadual Paulista, Instituto de Química Orientador: Wagner Vilegas Coorientador: Wagner Cotroni Valenti

1. Fauna marinha. 2. Produtos naturais. 3. Invertebrados marinhos. 4. Bactérias. 5. Actinobactéria. I. Título.

Elaboração: Seção Técnica de Aquisição e Tratamento da Informação Biblioteca do Instituto de Química, Unesp, câmpus de Araraquara

SÃO PAULO STATE UNIVERSITY “Júlio de Mesquita Filho” Institute of Chemistry - Campus Araraquara

Marcelo Marucci Pereira Tangerina

By-catch: A Chemical Universe To Be Explored

Doctoral Thesis presented as part of the requirements for obtaining the title of Doctor in Chemistry.

Supervisor: Prof. Dr. Wagner Vilegas Co-Supervisor: Prof. Dr. Wagner C. Valenti

Araraquara 2016

Dados Curriculares

Dados Pessoais

Nome: Marcelo Marucci Pereira Tangerina

Endereço Profissional:

1) Universidade Estadual Paulista “Júlio de Mesquita Filho” - UNESP, Instituto de Química, Departamento de Química Orgânica, Laboratório de Fitoquímica Rua Prof. Francisco Degni, 55. Jd. Quitandinha, CEP: 14800-900 – Araraquara – SP. Telefone: 16 3301-9792

2) Universidade Estadual Paulista “Júlio de Mesquita Filho” - UNESP, Instituto de Biociências, Laboratório de Bioprospecção de Produtos Naturais Praça Infante Dom Henrique, s/n. Parque Bitaru CEP: 11330-900 – São Vicente – SP. Telefone: 13 3569-7169

E-mail: [email protected]; [email protected]

Formação Acadêmica

Graduação

Bacharelado

Instituição: Universidade Estadual Paulista “Júlio de Mesquita Filho” - UNESP, Instituto de Química, Campus Araraquara Local: Araraquara - SP Curso: Bacharelado em Química com Atribuições Tecnológicas Período: 2006-2009 Trabalho Orientado em Química – Departamento de Química Inorgânica: - Investigação de Compostos de Ni(II) Coordenado a Diaminas e Pseudohaletos – Bolsista CNPq - Investigação de compostos de Ni(II) coordenado a azida e aminoalcoois como precursores de filmes metálicos – Bolsista FAPESP Orientadora: Prof.ª Dr.ª Vânia Martins Nogueira

Pós Graduação

Mestrado

Instituição: Universidade Estadual Paulista “Júlio de Mesquita Filho” - UNESP, Instituto de Química, Departamento de Química Orgânica, Campus Araraquara Local: Araraquara - SP Período: 2010-2011 Bolsista: FAPESP – Processo no. 2010/04077-0 Título da Dissertação: Extratos Padronizados para o Tratamento de Doenças Crônicas: Machaerium hirtum

Orientador: Prof. Dr. Wagner Vilegas Co-Orientadora: Prof.ª Dr.ª Miriam Sannomiya

Doutorado

Instituição: Universidade Estadual Paulista “Júlio de Mesquita Filho” - UNESP, Instituto de Química, Departamento de Química Orgânica, Campus Araraquara Local: Araraquara – SP/ São Vicente – SP Período: 2012-2016 Bolsista: FAPESP – Processo no. 2011/23159-0 Título da Tese: Fauna Acompanhante: Um Universo Químico a Ser Explorado Orientador: Prof. Dr. Wagner Vilegas Co-Orientador: Prof.ª Dr.ª Wagner C. Valenti

Estágio em Instituição de Ensino no Exterior:

1) Instituição: University of Regensburg – UR Local: Regensburg, Alemanha Período: Agosto/2013 – Dezembro/2013 Bolsista: CAPES Projeto: Diversidade e ecologia de caranguejos de manguezal do sudeste do Brasil exemplificado pelas comparações genéticas, morfológicas, etológicas, fisiológicas e ecotoxicológicas de duas espécies intimamente relacionadas de caranguejos violinistas. Orientador no Exterior: Prof. Dr. Christoph D. Schubart Orientador no Brasil: Prof.a Dr.a Tânia Márcia Costa

2) Instituição: University of Prince Edward Island – UPEI Local: Charlottetown, PE, Canadá Período: Julho/2014 – Março/2015 Bolsista: FAPESP (BEPE – Processo no. 2014/08787-2) Projeto: Fauna Acompanhante: Um universo Químico a Ser Explorado Trabalho realizado: Isolamento de bactérias marinhas e bioprospecção das mesmas Orientador no Exterior: Prof. Dr. Russell G. Kerr Orientador no Brasil: Prof. Dr. Wagner Vilegas

“Aos meus pais que sempre me apoiaram em todas as decisões que tomei até aqui e à toda minha família, dedico este trabalho.”

Agradecimentos

Agradeço aos meus pais, minha avó e meu irmão por ser quem são, pessoas excepcionais que sempre me apoiaram, compartilharam minhas alegrias, minhas angústias e se fizeram presentes por todo caminho que trilhei. Ao meu orientador, Wagner Vilegas, que formou o profissional que sou hoje. Sempre será meu orientador, o qual respeito muito, mas considero também um grande amigo que ganhei durante minha pós-graduação e hoje tenho em grande estima. Aos meus colegas e amigos do laboratório Claudinha, Douglas, Polly, Mayara, Marcelo, Julia, Lucas, Luiza, Nis, pela ajuda no laboratório, conversas informais, desabafos, cafés e todas as confraternizações. Aos meus grandes amigos que conheço desde minha graduação Jerso, Miguel, Glauco, Mari e Fabi pelas risadas e amizade verdadeira. Ao bando de biólogos que conheci durante o período no Campus de São Vicente da UNESP, Pri, Kiwi, Tixa, Dona, Shan, Ana, Cala, Bartira, Vem, Fer e a tantos outros aqui não citados. Aos professores da UNESP de São Vicente, Wagner C. Valenti e Tânia Márcia Costa e à professora do Instituto de Química de Araraquara, Lourdes Campaner pelas oportunidades e ótimo convívio. Ao Prof. Russell Kerr pela oportunidade de estágio em seu laboratório na University of Prince Edward Island – Canada e a todos que me ajudaram, em especial Hebelin, Brad e Fabrice. À Rede SAO-MAR/CNPq. À Capes pelo financiamento do estágio no exterior na Alemanha. À FAPESP pelas bolsas concedidas no país (Processo no. 2011/23159-0) e no exterior (BEPE - Processo no. 2014/08787-2).

“It must be a strange world not being a scientist, going through life not knowing – or maybe not caring about where the air came from, where the stars at night came from or how far they are from us. I Want To Know.” - Michio Kaku

Resumo Expandido

O desperdício de recursos provenientes da biodiversidade é enorme no Brasil e no mundo. A grande variedade de organismos marinhos (algas, moluscos, esponjas, corais, etc) que são pescados em conjunto com peixes e camarões, chamada de fauna acompanhante, é completamente ignorada como fonte de novas substâncias. Devido ao baixo valor agregado, os pescadores os desprezam, pois não são comercializáveis como alimento. A pesca de camarões é uma das mais impactantes aos ecossistemas marinhos: o uso de sistemas de arrasto do fundo marinho provoca a destruição de todo a região onde ele é empregado, capturando até 21 kg de fauna acompanhante/kg de camarão. Diversas substâncias encontradas em organismos marinhos são extensamente aplicadas em diversas áreas para a melhoria da qualidade de vida da humanidade, incluindo a área de alimentos, de geração de energia e medicamentos. Tendo em vista esse quadro, este projeto visou o estudo químico de espécies da fauna acompanhante da pesca do camarão no litoral paulista, visando obter substâncias de potencial interesse econômico, que possam aumentar o valor agregado desse material desperdiçado. Para o desenvolvimento do trabalho foram realizados três arrastos de fundo nas regiões de Ubatuba, Guarujá e Itanhaém, no estado de São Paulo, juntamente aos pescadores camaroeiros locais. O material coletado foi levado ao Campus de São Vicente da UNESP, onde foi triado e as espécies coletadas devidamente identificadas. Um total de 23 taxa de invertebrados (entre crustáceos, moluscos, cnidários e equinodermos), 07 elasmobrânquios (peixes cartilaginosos) e 64 peixes ósseos foram identificados, sendo os invertebrados identificados pela Prof.a Dr.a Tânia M. Costa, e os vertebrados pelos Prof. Dr. Otto Gadig (elasmobrânquios) e Prof. Dr. Teodoro Vasquez (peixes ósseos), docentes da UNESP/CLP. Esta parte do trabalho gerou o artigo intitulado “Biological and preliminary chemical characterization of the by-catch of the shrimp fishery from the São Paulo State coast, Brazil”, submetido ao periódico Latin American Journal of Aquatic Research Os peixes ósseos e os elasmobrânquios, por não ser foco do trabalho, não foram avaliados quimicamente. Quanto aos invertebrados, apesar de uma grande variedade de espécies, a primeira dificuldade encontrada foi a quantidade de indivíduos coletados por espécie. Assim, a maioria dos invertebrados coletados não apresentou massa suficiente para estudos aprofundados de composição química. Dentre as espécies de invertebrados que apresentaram uma maior quantidade de material, foram estudadas as águas-vivas Olindia

sambaquiensis, Chrysaora lactea e Chiropsalmus quadrumanus, os gastrópodes Olivancillaria urceus e Buccinanops cochlidium, os cefalópodes Lolliguncula brevis e Dorytheutis plei e o equinodermo Luidia senegalensis. Entretanto, excetuando a espécie de equinodermo, as demais apresentaram apenas metabólitos primários nos estudos realizados. O estudo de L. senegalensis foi possível por meio de espectrometria de massas uma vez que não foi capturado um número elevado de indivíduos. Para tanto, foi utilizada uma combinação de extração em fase sólida (SPE) seguida de cromatografia líquida de ultra eficiência acoplada a um espectrômetro de massas com ionização por electrospray e analisador íon-trap linear (UPLC-ESI-IT-MSn). Foi também realizada análise por injeção direta utilizando a mesma fonte de ionização e mesmo analisador, com fragmentação sequencial dos íons detectados (FIA-ESI-IT-MSn). A espécie apresentou asterosaponinas, as quais são esteroides glicosilados sulfatados, contendo cinco ou seis unidades de açúcar, além de poliidroxiesteroides. Tais resultados evidenciaram a presença de importantes substâncias potencialmente bioativas em invertebrados provenientes da fauna acompanhante da pesca do camarão utilizando um método rápido e eficiente. Esta etapa resultou no artigo “Chemical profile of the sulphated saponins from the starfish Luidia senegalensis collected as by-catch fauna in Brazilian coast”, submetido ao periódico Journal of the Brazilian Chemical Society. Devido à baixa abundância de produtos naturais detectados nos invertebrados provenientes da fauna acompanhante, a busca por novas substâncias foi ainda estendida às espécies microbianas associadas ao material de estudo. Assim, foram também estudadas as bactérias marinhas associadas aos invertebrados coletados e ao sedimento marinho do mesmo local de coleta quanto ao seu potencial na produção de substâncias. Foi explorado o potencial biotecnológico das bactérias cultiváveis de duas espécies de invertebrados da fauna acompanhante, o gastrópode Olivancillaria urceus e a estrela-do-mar L. senegalensis. Foi também estudada uma amostra de sedimento marinho proveniente da mesma área de coleta dos invertebrados. Utilizando múltiplas técnicas de isolamento foram obtidos 134 isolados bacterianos dos invertebrados e do sedimento. O sequenciamento parcial da subunidade do gene rRNA (16S) revelou que os isolados pertencem as filos Proteobacteria, Firmicutes e Actinobacteria, distribuídos em 28 gêneros. Diversos gêneros conhecidos por sua capacidade de produção de substâncias bioativas (Micromonospora, Streptomyces, Serinicoccus e Verrucosispora) foram obtidos das amostras estudadas. A fim de investigar as bactérias quanto a sua capacidade de produção de metabólitos bioativos, os isolados foram cultivados e o caldo de fermentação analisado por cromatografia líquida de ultra eficiência acoplada a detector de arranjo de fotodiodos, detector evaporativo de espalhamento de luz e

espectrômetro de massas de alta resolução com ionização por electrospray e analisador orbitrap (UPLC-PDA-ELSD-HRMS) e testados por sua atividade antimicrobiana. Por fim, quatro cepas apresentaram atividade antimicrobiana contra Staphylococcus aureus resistente à meticilina (MRSA) e Staphylococcus warneri. Esta etapa resultou no artigo intitulado “Cultivable bacterial communities of marine sediment and invertebrates from the underexplored Ubatuba region of Brazil”, submetido ao periódico Archives of Microbiology. A produção de substâncias por bactérias é altamente dependente do meio de cultura no qual estas são cultivadas. Ademais, diferentes cepas da mesma espécie podem apresentar diferente produção de metabólitos. A fim de avaliar tais variáveis foi realizado o cultivo em pequena escala em nove meios de cultura diferentes de três cepas da espécie Verrucosispora maris, provenientes do sedimento marinho e de L. senegalensis. Devido ao grande número de amostras obtidas, foi realizado um estudo metabolômico utilizando LC-HRMS e análise dos componentes principais (PCA), onde foi investigada a produção de abyssomicinas (marcadores químicos do gênero que apresentam propriedades anticancerígenas) e outros metabólitos secundários. Foi possível detectar a produção de abyssomicinas somente por uma das cepas avaliadas (RKMT_111) e o estudo da influência da composição do meio de cultura na produção de substâncias revelou que tais metabólitos só foram produzidos em um dos meios de cultura (BFM-11m). Apesar das três cepas pertencerem à mesma espécie e serem provenientes do mesmo local de coleta, é notável que todas apresentaram diferente capacidade de produção de metabólitos secundários. Tais resultados evidenciam a importância de uma triagem prévia dentre cepas de uma determinada espécie e da otimização da composição do meio de cultura a ser utilizado antes de fermentações em larga escala para a produção e posterior isolamento de substâncias provenientes de bactérias. Esta etapa do trabalho resultou no artigo intitulado “Survey of the secondary metabolome of the marine actinomycete Verrucosispora sp”, a ser submetido no periódico Molecules. Além da produção de metabólitos secundários as bactérias podem ainda realizar biotransformações, ou seja, modificações na estrutura de substâncias modificando suas propriedades físico-químicas. A fim de detectar tais transformações, foram estudados os produtos da fermentação de Erythrobacter vulgaris, isolado de sedimento marinho. Para tanto, o isolado foi fermentado em escala ampliada, o caldo de fermentação extraído e as substâncias isoladas por meio de cromatografia líquida preparativa utilizando detector de massas (HPLC-MS). As frações obtidas foram analisadas por UPLC-PDA-ELSD-HRMS e Ressonância Magnética Nuclear (RMN) e foram identificados dois novos derivados do ácido cólico, o ácido 3-acetil-glicocólico e o ácido 3-acetil-glicodesoxicólico. Sugere-se que as duas

novas substâncias isoladas foram produzidas por meio da biotransformação dos ácidos glicocólico e glicodesoxicólico, respectivamente, previamente presentes no meio de cultivo. Este é o primeiro registro dos compostos identificados bem como o primeiro estudo em que foi observada uma acilação realizada por um isolado marinho de Erythrobacter vulgaris. Estes resultados geraram o artigo intitulado “New glycocholic and glycodeoxycholic acid derivatives produced by biotransformation in Erythrobacter vulgaris marine isolate”, a ser submetido no periódico Journal of Biotechnology. A busca por novos compostos que sirvam de inspiração para o desenvolvimento de substâncias úteis para a humanidade é cada vez mais difícil e novas fontes devem ser buscadas a cada estudo. A fauna acompanhante mostrou ser uma fonte rica e ainda inexplorada na busca de bactérias marinhas com diversas aplicações biotecnológicas, ainda a serem exploradas. Quanto à composição química dos invertebrados estudados, a fauna acompanhante mostrou-se pouco promissora devido à baixa abundância de cada espécie e à grande captura de animais pelágicos, os quais apresentam outros mecanismos de defesa que não sejam a produção de substâncias químicas.

Resumo

A fauna acompanhante da pesca do camarão inclui uma série de invertebrados marinhos que são descartados por não ter valor comercial. A fim de tentar acrescentar algum valor a este material, foi analisada a composição química da estrela-do-mar Luidia senegalensis coletada na costa brasileira como consequência da aplicação da pesca de arrasto. A fim de avaliar sua composição química, foi utilizada uma combinação de extração em fase sólida (SPE) seguida de cromatografia líquida de ultra eficiência acoplada a espectrômetro de massas equipado com fonte de ionização por eletrosptray e analisador ion-trap linear (UPLC- ESI-IT-MSn). Luidia senegalensis contém asterosaponinas, que são esteroides glicosilados sulfatados contendo cinco e seis unidades de açúcar, além de poliidroxiesteroides. Este estudo mostrou a presença de compostos importantes e potencialmente bioativos em invertebrados associados à fauna acompanhante da pesca do camarão, usando um método rápido e eficiente. Normalmente descartada, a fauna acompanhante contém muitos invertebrados que podem hospedar uma grande variedade de gêneros de bactérias, algumas das quais com potencial de produzir produtos naturais bioativos com aplicações biotecnológicas. Assim, para utilizar um material normalmente descartado, foi explorado o potencial biotecnológico de bactérias cultiváveis de duas espécies de invertebrados abundantes na fauna acompanhante, o gastrópode Olivancillaria urceus e a estrela-do-mar Luidia senegalensis. Uma amostra de sedimento da mesma área de coleta também foi investigado. Utilizando múltiplas abordagens de isolamento 134 isolados foram obtidos a partir dos invertebrados e do sedimento. Sequenciamento parcial da subunidade de rRNA (16S) revelou que os isolados pertenciam aos filos Proteobacteria, Firmicutes e Actinobacteria, distribuídos em 28 gêneros. Vários gêneros conhecidos pela sua capacidade de produzir produtos naturais bioativos (Micromonospora, Streptomyces, Serinicoccus e Verrucosispora) foram obtidos a partir das amostras estudadas. Para avaliar as bactérias isoladas quanto à sua capacidade para produzir metabólitos bioativos todas as cepas foram fermentadas e os extratos de fermentação analisados por LC-HRMS e testados em ensaio de atividade antimicrobiana. Quatro cepas apresentaram atividade antimicrobiana contra Staphylococcus aureus resistente à meticilina (MRSA) e Staphylococcus warneri. A produção de metabólitos secundários por bactérias isoladas da fauna acompanhante também foi avaliada por uma abordagem metabolômica utilizando LC-HRMS, onde foi avaliado como as diferenças na composição dos meios de cultura podem alterar a produção de

substâncias. Utilizou-se a metabolômica como uma ferramenta para investigar a produção de abyssomicinas, um agente anticâncer, e outros metabólitos secundários em três cepas do actinomiceto raro Verrucosispora maris, isoladas a partir de uma amostra de sedimento e associadas à estrela-do-mar Luidia senegalensis de Ubatuba - SP, Brasil. Nove composições diferentes de meios de cultura foram avaliadas e verificou-se que, dentre todas as cepas, somente RKMT_111 foi capaz de produzir abyssomicinas. O estudo da composição do meio de cultura revelou que a produção de abyssomicinas só foi possível em BFM-11m. Embora as três cepas pertençam à mesma espécie e são provenientes da mesma localização, é notável que cada isolado apresentou diferente capacidade de produção de metabólitos secundários. Os produtos de fermentação de Erythrobacter vulgaris foram avaliados utilizando técnicas de HPLC preparativo, LC-HRMS e RMN. A cepa foi isolada pelo método dry-stamp de uma amostra de sedimento marinho da costa de Ubatuba-SP, Brasil. Depois de sequenciamento completo do rRNA (16S) e identificação, o isolado foi fermentado em larga escala, seu caldo de fermentação extraído por solvente e os compostos purificados por HPLC- MS. Análise de LC-HRMS e RMN dos compostos isolados levou à identificação de dois novos derivados do ácido cólico, ácido 3-acetil-glicocólico e o ácido 3-acetil- glicodesoxicólico. As substâncias obtidas podem ter sido produzidas por biotransformação do ácido glicocólico e ácido desoxicólico, respectivamente, já presentes no meio de cultivo. Este é o primeiro relato de tais compostos e também a primeira observação de uma acilação realizada por um isolado marinho de Erythrobacter vulgaris.

Palavras-chave: Fauna acompanhante; produtos naturais marinhos; invertebrados; bactérias

Abstract

The by-catch fauna of the shrimp fishery includes a number of marine invertebrates that are discarded because they do not have commercial value. In order to try to add some value to these materials, we analyzed the chemical composition of the starfish Luidia senegalensis collected in the Brazilian coast as a consequence of the trawling fishery method. In order to access their chemical composition, we used a combination of solid phase extraction (SPE) followed by ultra performance liquid chromatography coupled to electrospray ionization ion trap tandem mass spectrometry (UPLC-ESI-IT-MSn). Luidia senegalensis contains asterosaponins, which are sulphated glycosilated steroids, containing five and six sugar moieties, in addition to polyhydroxysteroids. This study helped us to support the presence of important and potentially bioactive compounds in invertebrates associated to the by-catch fauna of the shrimp fishery, using a fast and efficient method. Typically discarded, by-catch contains many invertebrates that may host a great variety of bacterial genera, some of which may produce bioactive natural products with biotechnological applications. Therefore, to utilize by-catch that is usually discarded we explored the biotechnological potential of culturable bacteria of two abundant by-catch invertebrate species, the snail Olivancillaria urceus and the sea star Luidia senegalensis. Sediment from the collection area was also investigated. Utilizing multiple isolation approaches 134 isolates were obtained from the invertebrates and sediment. Small subunit rRNA (16S) gene sequencing revealed that the isolates belonged to Proteobacteria, Firmicutes and Actinobacteria phyla and were distributed among 28 genera. Several genera known for their capacity to produce bioactive natural products (Micromonospora, Streptomyces, Serinicoccus and Verrucosispora) were retrieved from the invertebrate samples. To query the bacterial isolates for their ability to produce bioactive metabolites all strains were fermented and fermentation extracts profiled by LC-HRMS and tested for antimicrobial activity. Four strains exhibited antimicrobial activity against methicillin-resistant Staphylococcus aureus (MRSA) and Staphylococcus warneri. The production of secondary metabolites was assessed using a LC-HRMS-based metabolomics approach, where it was evaluated how differences in media composition can alter the production of chemical compounds. We used metabolomics as a tool to investigate the production of abyssomicins, an anticancer agent, and other secondary metabolites in three strains of the rare actinomycete Verrucosispora maris, all marine isolates from a sediment

sample and associated to a starfish from the species Luidia senegalensis of Ubatuba – SP, Brazil. Nine different media compositions were evaluated and it was found that, among all strains, only RKMT_111 was capable of producing abyssomicins. The media composition study revealed that the production of abyssomicins was only achievable in BFM-11m. Although the three strains belong to the same species and the same location, it is worthwhile noticing that each isolate showed different capability for production of secondary metabolites. The products of fermentation of Erythrobacter vulgaris were evaluated using preparative HPLC, LC-HRMS and NMR techniques. Bacterial strain was isolated by dry- stamp method from a marine sediment sample from the coast of Ubatuba-SP, Brazil. After fully 16S rDNA sequence and identification, the marine isolate was fermented in large-scale, extracted and the compounds purified through HPLC-MS. Analysis of LC-HRMS and NMR of the isolated compounds led to the identification of two new cholic acid derivatives, 3- acetyl-glycocholic acid and 3-acetyl-glycodeoxycholic acid. Both new compounds may have been produced by the biotransformation of glycocholic acid and deoxycholic acid, respectively, already present in the cultivation medium. This is the first report of such compounds and also the first time an acylation has been observed for an Erythrobacter vulgaris marine isolate.

Keywords: By-catch; marine natural products; invertebrates; bacteria

Lista de Figuras

Figure 1. Full scan DFI-ESI-IT-MSn (Negative Ionization) mass spectrum of Luidia senegalensis showing the peaks corresponding to the saponins ...... 48

Figure 2. UPLC-ESI-IT-MS analysis of the saponins present in Luidia senegalensis. Base Peak Ion -BPI (above) and extracted chromatograms of the ions 1: m/z 1405; 2: m/z 1389; 4: m/z 1239; 3 and 5: 1227; 6: m/z 529; 7: m/z 547 ...... 49

Figure 3. Aglycones of the steroidal saponins detected in Luidia senegalensis. R corresponds to the sugar sequence described in Table 2 ...... 50

Figure 4. Mass spectrum obtained after UPLC-ESI-IT-MSn experiment using the precursor ion ...... 52

Figure 5. Phylogenetic analysis of Proteobacteria isolates obtained from Brazilian sediment and the invertebrates Luidia senegalensis and Olivancillaria urceus. Prior to tree construction sequences sharing >99% sequence identity were grouped into OTUs and a single representative of each OTU was used in the phylogenetic analysis. The sources of isolates belonging to an OTU are indicated by symbols which follow the strain number: wet sediment - n; dry sediment - t; Luidia senegalensis - p; Olivancillaria urceus - l. Neighbor-joining tree constructed using MEGA 6. The analysis considered 577 nucleotides. Bootstrap values greater than 50% are shown at the nodes and are based on 1000 iterations. The scale bar represents the number of base substitutions per site ...... 58

Figure 6. Phylogenetic analysis of Firmicutes isolates obtained from Brazilian sediment and the invertebrates Luidia senegalensis and Olivancillaria urceus. Prior to tree construction sequences sharing >99% sequence identity were grouped into OTUs and a single representative of each OTU was used in the phylogenetic analysis. The sources of isolates belonging to an OTU are indicated by symbols which follow the strain number: wet sediment - n; dry sediment - t; Luidia senegalensis - p; Olivancillaria urceus - l. Neighbor-joining tree constructed using MEGA 6. The analysis considered 573 nucleotides. Bootstrap values greater than 50% are shown at the nodes and are based on 1000 iterations. The scale bar represents the number of base substitutions per site ...... 59

Figure 7. Phylogenetic analysis of Actinobacteria isolates obtained from Brazilian sediment and the invertebrates Luidia senegalensis and Olivancillaria urceus. Prior to tree construction sequences sharing >99% sequence identity were grouped into OTUs and a single representative of each OTU was used in the phylogenetic analysis. The sources of isolates belonging to an OTU are indicated by symbols which follow the strain number: wet sediment - n; dry sediment - t; Luidia senegalensis - p; Olivancillaria urceus - l. Neighbor-joining tree constructed using MEGA 6. The analysis considered 621 nucleotides. Bootstrap values greater than 50% are shown at the nodes and are based on 1000 iterations. The scale bar represents the number of base substitutions per site ...... 60

Figure 8. Effect of media composition in the production of secondary metabolites of three strains of Verrucosispora maris. Principal component analysis (PCA) – Scores plot (left) and loadings plot (right), PC-1 versus PC-2. (A) Strain RKMT_073; (B) Strain RKMT_111; (C) Strain RKMT_176; (D) All three strains analyzed together ...... 68

Figure 9. Chemical barcoding and cluster analysis of the three strains of Verrucosispora maris in all media tested ...... 74

Figure 10. Phylogenetic analysis of RKMT_070 isolate obtained from Brazilian sediment and reference strains from Genbank. Neighbor-joining tree constructed using MEGA 6. The analysis considered 1336 nucleotides. Bootstrap values greater than 50% are shown at the nodes and are based on 1000 iterations. The scale bar represents the number of base substitutions per site ...... 75

Figure 11. Structure of compounds isolated from Erythrobacter vulgaris fermentation ...... 78

Figure S1. Chromatographic profile of the extract from Streptomyces sp. strain RKMT_071. ELSD, HRMS and PDA detectors. Below, extracted chromatograms of the ions of m/z 479.29080 [M+H]+, 1111.64453 [M+H]+, 1147.64099 [M+Na]+ and 560.31848 [M+H]+. .. 101

Figure S2. UV profile of the peak at retention time of 4.54 min in the PDA detector from the analysis of the extract from RKMT_071 strain ...... 101

Figure S3. Chromatographic profile of the extract from Micromonospora sp. strain RKMT_160. ELSD, PDA and HRMS detectors. Below, extracted chromatograms of the ions of m/z 261.12335 [M+H]+ and 284.13940 [M+H]+ ...... 102

Figure S4. Chromatographic profile of the extract from Bacillus sp. strain RKMT_178. ELSD, PDA and HRMS detectors. Below, extracted chromatograms of the ions of m/z 345.18419 [M+H]+ and 389.19112 [M+H]+ ...... 102

Figure S5. Chromatographic profile of the extract from Halobacillus sp. strain RKMT_184. ELSD, PDA and HRMS detectors ...... 103

Figure S6. Chromatographic profile of the extract from Halobacillus sp. strain RKMT_184. ELSD and HRMS detectors. Below, extracted chromatograms of the ions of m/z 316.12922 [M+H]+, 351.12129 [M+H]+, 325.12518 [M+H]+, 217.09728 [M+H]+, 349.17912 [M+H]+ and 375.17545 [M+H]+ ...... 103

Figure S7. Extracted chromatograms in SIM mode for the ion [M+H]+ of m/z 231.1016 from RKMT_111 (top) and RKMT_176 (bottom) in BFM-11m corresponding to the compound Methyl 3-methoxy-5-methyl-naphthalene-1-carboxylate ...... 104 Figure S8. Extracted chromatograms in SIM mode for the ion [M+H]+ of m/z 257.1172 corresponding to the compound kurasoin A. A – RKMT_111 in BFM-1m; B – RKMT_176 in BFM-1m; C – RKMT_111 in BFM-11m; D – RKMT_176 in BFM-11m...... 104

Figure S9. Extracted chromatograms in SIM mode for the ion [M+H]+ of m/z 247.1082 from RKMT_111 (top) and RKMT_176 (bottom) in BFM-11m corresponding to the compound anthramycin ...... 105

Figure S10. Extracted chromatograms in SIM mode for the ion [M+H]+ of m/z 223.0964 from RKMT_111 (top) and RKMT_176 (bottom) in BFM-11m corresponding to the compound talomone ...... 105

Figure S11. Extracted chromatograms in SIM mode for the ion [M+H]+ of m/z 219.1016 from RKMT_111 (top) and RKMT_176 (bottom) in BFM-11m corresponding to the compound 4H-1,3-Benzodioxin-4-one, 2,2-dimethyl-5-(2-propen-1-yl)- ...... 106

Figure S12. Extracted chromatograms in SIM mode for the ion [M+H]+ of m/z 293.1383 from RKMT_111 (top) and RKMT_176 (bottom) in BFM-11m corresponding to the compound 4H-1-Benzopyran-4-one, 2-butyl-8-hydroxy-5,7-dimethoxy-3-methyl- ...... 106

Figure S13. Extracted chromatograms in SIM mode for the ion [M+H]+ of m/z 521.3475 from RKMT_111 (top) and RKMT_176 (bottom) in BFM-4m corresponding to the compound butyrolactol ...... 107

Figure S14. Comparison of the composition of all fractions obtained by extraction with ethyl acetate and HP-20 of the broth from cultivation of RKMT_070 in tubes and baffled flasks. Left: ELSD detector; Center: PDA detector; Right: HRMS detector. A to D: cultivation in baffled flasks, ethyl-acetate, HP-20 H2O, HP-20 MeOH/H2O, HP-20 MeOH, respectively. E to H: cultivation in tubes, ethyl-acetate, HP-20 H2O, HP-20 MeOH/H2O, HP-20 MeOH, respectively...... 117

Figure S15. Chromatogram of preparative HPLC-MS of fraction HP-20 MeOH. Sunfire C18 column (5 µm, 250 × 10 mm, 110 Å, Waters®). Method described in Isolation section. Numbers in the chromatogram correspond to the method fractions were collected ...... 117

Figure S16. Compounds detected in the medium blank analysis by LC-HRMS. A: Base Peak chromatogram of the medium blank. Extracted ion chromatograms of the ions B: m/z 466.4; C: m/z 407.3; D: m/z 409.3; E: m/z 450.3 ...... 118

Figure S17. Extraction of the ions of m/z 492.3 (B) and m/z 508.3 (C) from analysis of the medium blank (A), confirming the absence of the isolated compounds in the medium composition ...... 118

Figure S18. 1H-NMR spectrum of Fraction 07 (14.0 T, MeOD, ppm) ...... 119

Figure S19. HMQC contour map of Fraction 07 (14.0 T, MeOD, ppm) ...... 119

Figure S20. HMBC contour map of Fraction 07 (14.0 T, MeOD, ppm) ...... 120

Figure S21. 1H-NMR spectrum of Fraction 11 (14.0 T, MeOD, ppm) ...... 120

Figure S22. HMQC contour map of Fraction 11 (14.0 T, MeOD, ppm) ...... 121

Figure S23. HMBC contour map of Fraction 11 (14.0 T, MeOD, ppm) ...... 121

Lista de Tabelas

Table 1. Species belonging to Verrucosispora genus, location and source of first report ...... 28

Table 2. Proposed saponins present in Luidia senegalensis detected by ESI-IT-MSn (1), UPLC-ESI-IT-MS and UPLC-ESI-IT-MSn (2) in negative ion mode ...... 50

Table 3. Summary of isolates obtained from different samples using different pretreatments and isolation media. Numbers in represent “number of isolates/number of genera” ...... 57

Table 6. Summary of metabolites observed in UPLC-HRMS analyses of fermentation extracts from four strains. To identify metabolites detected in these analyses, the pseudomolecular (PM) ion m/z was used to search the Antibase 2014 database using a 5 ppm window above and below the observed molecular weight (MW). A 10 ppm search window was used for pseudomolecular ions with m/z >1000. Compounds with no matches in Antibase were considered putatively novel ...... 66

Table 7. Chromatographic data of the fractionation and analysis by LC-HRMS of the fractions ...... 77

Table 8. Chemical shifts of glycocholic, glycodeoxycholic (IJARE et al., 2005), 3-acetyl- glycocholic and 3-acetyl-glycodeoxycholic acids (MeOD) ...... 79

Sumário

1. Introduction ...... 23 1.1. The by-catch fauna and marine invertebrates ...... 23 1.2. Marine bacteria and its biotechnological potential ...... 25 1.3. Metabolomics and the chemical study of marine bacteria ...... 26 1.4. Proteobacteria and biotransformations ...... 29

2. General and Specific Goals ...... 31

3. Materials and Methods ...... 32 3.1. Mass spectrometry study of the starfish Luidia senegalensis ...... 32 3.1.1. Chemicals and Materials ...... 32 3.1.2. Animal material, extraction and fractionation ...... 32 3.1.3. ESI-MSn analysis ...... 33 3.1.4. UPLC-ESI-IT-MS analysis ...... 34 3.2. Marine bacteria ...... 34 3.2.1. Cultivable bacterial communities of marine sediment and invertebrates ...... 34 3.2.1.1. Sample collection ...... 34 3.2.1.2. Methods for the isolation of bacteria ...... 35 3.2.1.3. Culture media for bacterial isolation ...... 36 3.2.1.4. Bacteria identification ...... 37 3.2.1.5. Phylogenetic analysis of cultured bacteria ...... 38 3.2.1.6. Bacteria cultivation and extraction ...... 38 3.2.1.7. Antimicrobial assays ...... 39 3.2.1.8. UPLC-HRMS analysis ...... 39 3.2.2. Metabolomic study of the marine actinomycete Verrucosispora sp...... 40 3.2.2.1. Bacteria isolation and identification ...... 40 3.2.2.2. Bacteria fermentation ...... 40 3.2.2.3. Extraction and LC-HRMS analysis ...... 41 3.2.2.4. Preprocessing and statistical analysis ...... 42 3.2.2.5. Identification of compounds ...... 43 3.2.3. Biotransformations in bacterial isolates ...... 43 3.2.3.1. Bacteria isolation and identification ...... 43 3.2.3.2. Fermentation, extraction and LC-HRMS analysis ...... 44 3.2.3.3. Isolation and identification of compounds ...... 45

4. Results ...... 46 4.1. Mass spectrometry study of the starfish Luidia senegalensis…..………………...….46 4.1.1. SPE fractionation of the crude extract ...... 47 4.1.2. DFI-ESI-IT-MSn analyses ...... 47

4.1.3. UPLC-ESI-IT-MS analysis ...... 48 4.2. Marine bacteria ...... 54 4.2.1. Cultivable bacterial communities of marine sediment and invertebrates ...... 54 4.2.1.1. Bacteria isolation and identification ...... 54 4.2.1.2. Phylogenetic analysis ...... 58 4.2.1.3. Antimicrobial assay ...... 63 4.2.1.4. Chemical analysis of bioactive extracts ...... 63 4.2.2. Metabolomic study of the marine actinomycete Verrucosispora sp...... 66 4.2.2.1. Identification of bacterial strains ...... 66 4.2.2.2. Bacteria fermentation and metabolite production evaluation ...... 67 4.2.2.3. Metabolite production of the isolated strains ...... 67 4.2.2.4. Compilation of the data ...... 68 4.2.2.5. RKMT_073 metabolites production ...... 69 4.2.2.6. RKMT_111 metabolites production ...... 69 4.2.2.7. RKMT_176 metabolites production ...... 71 4.2.2.8. Comparison of metabolites production between strains ...... 72 4.2.2.9. Cluster analysis and barcode ...... 73 4.2.3. Biotransformations in bacterial isolates ...... 75 4.2.3.1. Identification of the bacterial strain ...... 75 4.2.3.2. Comparison of samples by LC-HRMS analysis ...... 76 4.2.3.3. Isolation and identification of compounds ...... 76 4.2.3.4. NMR analysis ...... 78

5. Discussion ...... 81 5.1. Mass spectrometry study of the starfish Luidia senegalensis….………..…………..81 5.2. Marine bacteria ...... 82 5.2.1. Cultivable bacterial communities of marine sediment and invertebrates ...... 82 5.2.2. Metabolomic study of the marine actinomycete Verrucosispora sp...... 85 5.2.3. Biotransformations in bacterial isolates ...... 87

6. Conclusion ...... 89

References ...... 90

Supplementary Material ...... 101

1. Introduction

1.1. The by-catch fauna and marine invertebrates

Current selective and rational fishing practices involve avoiding prohibited species and those with no commercial value. However, equipment required to catch high-value species such as fish and shrimp and the lack of robust markets for many unintentionally harvested marine organisms, commonly referred to as by-catch, makes the practical application of selective fishing practices challenging (HALL; ALVERSON; METUZALS, 2000). By-catch fauna is defined as any organism that is not the intended target of the harvest, including fishes, turtles, crustaceans, mollusks and other organisms. By-catch is a global problem, and the Food and Agriculture Organization (FAO) estimates that nearly 7 million tons of by-catch are discarded annually, an amount equivalent to 8% of the world’s marine fisheries. The by- catch associated with shrimp trawling in tropical waters is particularly egregious, accounting for 28% of all by-catch (EAYRS, 2007). Bottom trawling fishing practices not only leads to low harvest selectivity, but also the destruction of the neritic biological diversity, particularly in the demersal-benthic layer. In spite of high by-catch rates, the marketable portion of the catch is sufficiently profitable that current fishing practices remain profitable. Continued use of non-selective fishing practices reduce biodiversity as the majority of discarded by-catch perishes, thereby disrupting the ecological balance of fished areas (GRAÇA LOPES et al., 2002; SANTOS, 2007; SEVERINO-RODRIGUES; HEBLING; GRAÇA-LOPES, 2004). The broad variety of marine organisms that are caught together with fish and shrimp is completely ignored as a source of new molecules. Due to the low economic value, the fishermen despise them, because they are not marketable as food. However, several substances found in marine organisms can be widely applied in many areas to improve the quality of life of humanity, including the areas of food, power generation and medicine. Therefore, we investigate the chemical compounds present in the by-catch fauna of the shrimp fishery on the coast of State São Paulo, Brazil, in order to obtain molecules of potential economic interest, which can increase the value of this wasted material (CATTANI et al., 2011; CLUCAS, 1997). This approach is not the ideal since this predatory activity should be banned, but until now no other alternative for the shrimp fishery was found. This work does not intend to solve the by-catch problem, but rather use the material that is usually discarded. 24

To start our work on the chemical composition of the by-catch fauna of the shrimp fishery, we have investigated the polar extract of the nine-armed starfish Luidia senegalensis Lamark collected in the coast of the São Paulo State, Brazil. Luidia Forbes (1839, Luidiidae, Asteroidea: Paxillosida) are bottom sea stars, which live in sandy or muddy substrate. The genus includes 49 species that occur in tropical and subtropical shallow waters (XIAO et al., 2013). Luidia senegalensis [syn. Asterias senegalensis Lamarck (1816), Luidia marcgravii Steenstrup in Lutken (1859, synonym according to Perrier (1875)] occurs at depths of up to 40 meters alongside the coast of South America, including southern Brazil, as well as around the coasts of Florida, in the Caribbean Sea and the Gulf of Mexico (CLARCK; DOWNEY, 1992). Starfish (called also sea stars) are marine invertebrates widely recognized as source of natural products. They belong to the class Asteroidea, phylum Echinodermata. The chemical composition of a number of starfishes has been investigated using several chromatographic and spectrometric techniques. This class of Echinodermata is rich in free polyhydroxysteroids and two main groups of steroid glycosides: asterosaponins and the glycosides derived from the polyhydroxysteroids. The asterosaponins present a ∆9,11-3β,6α-steroidal core, with four rings, a sulphate group at C-3, one or two oxygenated carbons at the side chain, and four to six sugar moieties attached to C-6. Common saccharide residues are pentoses (xylose, arabinose), deoxyhexoses (quinovose, fucose), hexoses (glucose, galactose) and 6-deoxy- xylo-hex-4-ulose (DXHU). Therefore, the extracts produced from these marine organisms are often very complex mixtures of free and sulphated highly oxygenated compounds as well as their sodium salts. These sulphated steroid oligoglycosides usually have molecular weight higher than 1200 Da and may include isomeric compounds. These substances have a wide variety of pharmacological activities. Among them, they act as anti-viral, anti-bacterial, anti- inflammatory, anti-fungal, hemolytic, activate tubulin polymerization, inhibit tumour proliferation and possess immunomodulatory activities. They are also involved in physiological and chemical defense, interspecific chemical communication, digestion and reproduction (DATTA; TALAPATRA; SWARNAKAR, 2015; DONG et al., 2011). Recently, mass spectrometry coupled or not to liquid chromatography has been proved to be a powerful tool for the investigation of the saponins present in the polar extracts of sea stars, without the need of prior separation (DEMEYER et al., 2014; POPOV et al., 2014). In our case, the analysis of the polar hydroethanolic extract of L. senegalensis was accomplished using a combination of solid-phase extraction (SPE), direct flow injection- electrospray-ion trap tandem mass spectrometry (DFI-ESI-IT-MSn) and an ultra-high 25

performance liquid chromatography-electrospray-ion trap tandem mass spectrometry (UPLC- ESI-IT-MSn) method. Substances were tentatively identified using the typical fragmentation of the aglycone side chain, the characteristic losses of the sugar moieties and comparison with the fragmentation pattern described in the literature (DE MARINO et al., 2003; DEMEYER et al., 2014; DONG et al., 2011; MINALE et al., 1985; POPOV et al., 2014).

1.2. Marine bacteria and its biotechnological potential

While the value of by-catch as a food-source is limited, many benthic organisms harvested as by-catch have value as sources of secondary metabolites with a wide range of biological activities. Many invertebrates also harbor a plethora of microorganisms, which also have the potential to produce novel bioactive compounds (BLUNT et al., 2015). In some cases, symbiotic microorganisms are the true producers of metabolites first isolated from marine invertebrates (MINCER et al., 2002). Microorganisms are widely recognized as one of the most important sources of bioactive natural products, many of which have found utility as treatments for a variety of diseases such as infectious diseases, cancer, hypercolesterolemia etc.(FISCHBACH; WALSH, 2009; HAYGOOD et al., 1999; HONG et al., 2009; SPELLBERG et al., 2008). Despite the awareness of the biomedical potential of microbial natural products, the first natural product isolated from a microorganism discovered by Brazilian researchers was reported only in 2000. Among microbial natural products discovered in Brazil, only a small percentage (9%) were discovered from bacteria (IÓCA; ALLARD; BERLINCK, 2014). Thus bacteria from Brazilian habitats represent an underexplored resource. Antibiotic-resistant pathogens are an increasing health threat that makes the need to find new antibiotics particularly urgent (KIM; KSHETRIMAYUM; GOODFELLOW, 2011). More than two-thirds of clinically used antibiotics are natural products or their semisynthetic derivatives. Despite this need for new bioactive compounds, natural product discovery has declined in the last few years, in part because of the rediscovery of known compounds and the difficulty in finding new antibiotics. Recent efforts to reinvigorate the antibiotic discovery via bioprospecting from underexplored ecological niches, unexplored bacterial taxa, and even the genomes of well-studied bacteria have yielded novel antimicrobial natural products, whereas new screening strategies have begun to circumvent the time consuming problem of rediscovery (FISCHBACH; WALSH, 2009). 26

Most natural product antibiotics have been discovered from bacteria within the order Actinomycetales (commonly referred to as actinomycetes), which are common soil inhabitants. Though more than 50% of the microbial antibiotics discovered so far originate from actinomycetes, two genera (Streptomyces and Micromonospora) account for most of these compounds (PROCÓPIO et al., 2012; WAGMAN; WEINSTEIN, 1980). Recent explorations of marine environments have established marine bacteria, including marine actinomycetes, as a promising source of bioactive natural products (EL AMRAOUI et al., 2014). This fact is exemplified by the discoveries of the potent anticancer agent salinosporamide A and the novel antibiotic abyssomycin C from the marine actinomycetes Salinospora tropica and Verrucosispora sp., respectively (BISTER et al., 2004; NIEWERTH et al., 2014). Two marine species commonly harvested as by-catch in bottom trawling fishing practices are the starfish Luidia senegalensis (Lamark, 1816) (CLARCK; DOWNEY, 1992) and the gastropod Ollivancilaria urceus (Röding, 1798) (BOLTEN, 1906). To add potential value to these by-catch organisms and to expand the breadth of Brazilian research in the area of bacterial natural product discovery we set out to study the culturable bacteria associated with L. senegalensis and O. urceus for their ability to produce bioactive natural products. A marine sediment sample collected in the same area as the invertebrates was also examined as sediments are an established source of natural product-producing marine bacteria (DALISAY et al., 2013; GONTANG; FENICAL; JENSEN, 2007; SPONGA et al., 1999). Due to their proven track record as a rich source of novel natural products we focused our isolation methods on those selective for actinomycetes. Isolated bacteria were identified by 16S rRNA gene sequencing and fermentation extracts from selected isolates were screened for antimicrobial activity. The chemical composition of bioactive extracts was characterized by liquid chromatography-high-resolution mass spectrometry to identify compounds potentially responsible for the observed bioactivity.

1.3. Metabolomics and the chemical study of marine bacteria

The oceans covers 70% of earth’s surface and harbors most of the biodiversity of the planet (FENICAL; JENSEN, 2006), which can be related to a great molecular diversity of natural products found in animals, plants and microorganisms (KÖNIG et al., 2006). Such 27

environment presents a role set of factors that affect its inhabitants like pressure, salinity, temperature and nutrient availability (BOSE et al., 2015), which may affect their metabolism. Among all the living beings from the seas, microorganisms stand out for their capacity to thrive in several marine environments, from the water surface until the lower and abyssal depths, from coastal to offshore regions and from the general oceanic to the specialized niches (DAS; LYLA; KHAN, 2006). This fact shows the high adaptability of these organisms, due to their genetic plasticity and rapid replication, making microbes to be the most numerous, diverse and adaptable organisms on earth (SPELLBERG et al., 2008). This organisms have drawn the attention of scientists for many years due to their importance in many life processes, detrimentally or in the production of useful compounds such as vitamins, antibiotics and other pharmaceuticals (IÓCA; ALLARD; BERLINCK, 2014). In fact, almost 70% of small molecules that are utilized as medicines are derived or inspired in natural products produce by bacteria, more specifically filamentous actinobacteria (SEIPKE, 2015). When studying microorganisms, several parameters must be accessed to fully understand the production of secondary metabolites by one strain. For example, media composition, pH, temperature, oxygen availability and light intensity may affect microbes metabolism, therefore affecting compound production (RATEB et al., 2011). This parameters can be evaluated using the OSMAC (One Strain – Many Compounds) approach, where the alteration of culture conditions may turn on silent or cryptic biosynthetic genes and generate new metabolites where it is been reported from one single strain it could be isolated up to 20 different metabolites in yields up to 2.6 g L-1 (BODE et al., 2002; CHRISTIAN et al., 2005; RATEB et al., 2011). One of the major difficulties in natural products research is the rediscovery of known compounds. This problem can be avoided by the use of metabolomics, with hyphenated techniques to the fast and reliable evaluation of large sets of samples to prioritize the most promising sources of unknown compounds (HARVEY; EDRADA-EBEL; QUINN, 2015). The principal platforms utilized nowadays for metabolomics studies are LC-NMR and LC- MS assemblies, the second one being used in a larger proportion due to its lower cost and faster optimization and data processing (IBEKWE; AMEH, 2015). LC-MS based metabolomics is also a powerful tool in the study of known compounds production in different conditions (HARVEY; EDRADA-EBEL; QUINN, 2015). For example, seasonality studies in plant natural products (KWAK; HEGEMAN; PARK, 2014), chemical composition of marine invertebrates from different site collections and compound production of microorganisms (e.g. fungi, bacteria) submitted to different cultivation conditions (BOSE et 28

al., 2015). In this way, it is possible to find the best conditions for the production of bioactive compounds for better yields and time optimization. An interesting and rare genus of marine actinomycete, Verrucosispora, contains species that are a good example of promising producers of bioactive metabolites (FIEDLER et al., 2005). The genus comprises only eight species isolated from mangroves, summarized in Table 1.

Table 1. Species belonging to Verrucosispora genus, location and source of first report Species Location Source Ref Guangdong Province, Mangrove (LIAO et al., V. lutea China sediment 2009) V. qiuiae Hainan Province, China Swamp sediment (XI et al., 2012) V. Mangrove Wenchang, China (LIN; LI, 2012) wenchangensis sediment (RHEIMS et al., V. gifthornensis Gifhorn, Germany Peat sample 1998) Deep-sea sediment V. sediminis South China Sea (DAI et al., 2010) sample Deep-sea sediment (GOODFELLOW V. maris East Sea of Japan sample et al., 2012) (GOODFELLOW V. fiedleri Norway Fjord sediment et al., 2013) Phuket Province of Sponge (SUPONG et al., V. andamanensis Thailand Xestospongia sp. 2013)

Such organisms are great producer of bioactive compounds like the antibiotic and anticancer agents proximicins A, B and C (FIEDLER et al., 2008), produced by isolates of V. maris and V. fiedleri (GOODFELLOW et al., 2013; ROH et al., 2011), the diterpenes inhibitors of androgen receptors gifhornenolones A and B, produced by V. gifhornensis isolates (SHIRAI et al., 2010) and the potent antibiotic abyssomicins and its derivatives (BISTER et al., 2004; KELLER et al., 2007; WANG et al., 2013), from V. maris isolates and presenting a completely new scaffold, reinforcing the promising source of novel bioactive compounds from the marine environment. Also the cytotoxic thiodepsipeptide thiocoraline and its analogs were found to be produced by other Verrucosispora sp. isolates (WYCHE et al., 2011). Therefore, a study was carried out with three strains of Verrucosispora sp., using LC- HRMS based metabolomics for the detection and identification of secondary metabolite production. The three isolates were fermented in nine different media each and the resulting broth extracted with ethyl acetate and dried. For the evaluation of media composition effect in 29

compound production the extracts were analyzed by LC-HRMS followed by data processing and statistical analysis including cluster and principal component analysis (PCA). This methodology allows the rapid identification of compounds such as putatively new metabolites as well the fermentation conditions leading to good production yield.

1.4. Proteobacteria and biotransformations

Another important feature of the marine bacteria is the capability of biotransformation, providing new derivatives of known compounds with potential different activity, without the need of time-consuming low-yield synthesis steps (LI et al., 2006). Biotransformations by microorganisms are an important tool in industrial processes (BOAVENTURA; LOPES; TAKAHASHI, 2004; RODRÍGUEZ‐GARCÍA et al., 2016). Among its advantages are the less drastic conditions applied compared to synthesis like neutral pH, ambient temperatures and atmospheric pressure (HEGAZY et al., 2015) and the factor that are usually carried out in aqueous systems, which avoids the use of harmful solvents common in synthesis (BOAVENTURA; LOPES; TAKAHASHI, 2004). Moreover biotransformations are, in general, regio- and stereoselective (BAYDOUN et al., 2014), a major advantage compared to common synthesis. Several microorganisms can be used in processes for biotransformations, such as fungi (GAUTHIER et al., 2016; HEIDARY; HABIBI, 2016; KOLLEROV et al., 2015), microalgae (GHASEMI; RASOUL-AMINI; FOTOOH-ABADI, 2011), yeasts (GORETTI et al., 2009) and bacteria (terrestrial and marine) (COLQUHOUN et al., 1998; HYLEMON; HARDER, 1998; LI et al., 2006). Marine bacteria are a singular group which shows unique characteristics like the production of salt-tolerant enzymes, which may be very useful in industrial processes (DEBNATH; PAUL; BISEN, 2007). The main group of marine bacteria explored for its genetic machinery is the actinomycetes (COLQUHOUN et al., 1998; ISHIHARA et al., 2011), but Proteobacteria also contains individuals with potential applications in biotransformations (LI et al., 2006). In fact, reports say that marine Proteobacteria may have a higher complement of enzymes than Actinobacteria, differently from their terrestrial analogs where the opposite occurs (MUHLING; JOINT; WILLETTS, 2013). Therefore a study was carried out with a Proteobacteria marine isolate, Erythrobacter vulgaris, isolated from a marine sediment sample from the southeast coast of Brazil, using 30

LC-HRMS, cultivation, chromatographic purification and NMR spectroscopy to identify compounds produced or possible biotransformations performed by the isolated strain. 31

2. General and Specific Goals

The main goal of the study was to evaluate the chemical composition of marine invertebrate species from the by-catch fauna of shrimp fisheries in the São Paulo State. Also, investigate the marine bacteria associated to the invertebrates and sediment for its biotechnological potential. In brief, the specific goals of the study were: • Investigate the chemical composition of the starfish Luidia senegalensis in order to identify its secondary metabolites • Isolate and characterize the cultivable marine bacteria associated to the invertebrates and to the sediment collected • Assess the capability of the isolated bacteria concerning its biotechnological potential for compound production and/or biotransformations

3. Materials and Methods

For this study, three beaches located in the north, central and south of the São Paulo State traditionally used for the commercial shrimp fishing were selected: 1) North: Ubatuba, Itaguá Beach (23°26'12"S, 23°27'78"S and 45°00'76"W at 45°02'94"W), 2) Central: in Guaruja, Perequê Beach (23°46'45"S at 23°58'07"S and 46°08'42"W at 46°09'33") and 3) South: Itanhaém, Praia dos Pescadores Beach (24°11'39"S at 24°12'48"S and 46°46'44"W at 46°47'54"W). In each locality, two trawls were performed, one in January (hot season, average temperature 21oC) and one in July (cold season, average temperature 19oC), in 2013. Local vessels were used, equipped with a semi-ballon otter trawl net, 2.0 cm in the collecting sac and bagger of 14 mm between adjacent nodes, at a mean speed of two knots and depths ranging between 10 - 20 m in 30 min each trawl. The salinity remained at approximately 33.0. The collected animals were previously separated by large taxonomic groups, placed in plastic bags and stored in Styrofoam box with ice for transporting to the laboratory. The lab activities consisted of thawing the material, sorting, identifying at the lowest possible taxonomic level according to the literature for each major group identified, measurement and weighing of specimens. Materials from the different sampling sites were analyzed separately.

3.1. Mass spectrometry study of the starfish Luidia senegalensis

3.1.1. Chemicals and Materials

Methanol Chromasolv LC–MS grade, chloroform, 1-propanol, formic acid, ninhidrin, anysaldehyde and sulfuric acid were acquired from Sigma-Aldrich (São Paulo, Brazil). Ultrapure water was produced using a Milli-Q system (Millipore, Bedford, MA, USA).

3.1.2. Animal material, extraction and fractionation

Individuals of Luidia senegalensis were collected in the city of Ubatuba – SP, The specimens were kept in ice for transportation and frozen after arriving in the laboratory until extraction. The starfish was identified by Prof. Dr. Tania Marcia Costa, from UNESP - 33

Coastal campus of São Vicente. In addition, the specimens were photographed to afford a visual voucher. The animals were thawed to room temperature, separated by species, crushed with mortar and pestles. One gram of each animal was extracted by maceration with 10 mL of ethanol 70% (3 days). The extract was filtered and the solvent was removed under vacuum at 40 °C, using a rotary evaporator. The SPE cartridge (500 mg, Macherey-Nagel, Chromabond C18 ec, Düren, Germany) was first preconditioned by the consecutive passing of 5 mL of methanol and then 5 mL of pure water by gravity. The extract obtained was solubilized in water at concentration of 1.0 mg mL-1, filtered on filter paper to remove macro particles and further filtered through a 0.45 µm membrane of a PTFE filter. The sample was loaded to the cartridge and first eluted with 5 mL of water (to eliminate salts and free amino acids). The cartridge was eluted again with 5 mL of water/methanol (90:10, v/v, named hydromethanolic fraction - HF) and finally with 5 mL of pure methanol. All three fractions were analyzed using two thin layer chromatography plates (TLC, silicagel, 20 x 20 cm, 250 µm layer, UV fluorescence 254 nm, Whatman Ltd, Maidstone, England; Eluent: chloroform:methanol:n-propanol:water 100:11:11:27 v/v) and separately revealed with ninhidrin (to detect amino acids) and anisaldehyde/sulfuric acid mixture (to detect secondary metabolites). Fraction collected in water contained only amino acids and other impurities and HF concentrated the compounds of interest. HF was transferred into clean tubes and dried under nitrogen gas at room temperature. The sample was redissolved in pure methanol/water 50:50, v/v to a concentration of 5 ppm and analyzed by mass spectrometry.

3.1.3. ESI-MSn analysis

Direct flow infusion of the samples was performed on a Thermo Scientific LTQ XL linear ion trap analyzer equipped with an electrospray ionization (ESI) source, both in positive and negative modes (Thermo, San Jose, CA, USA). It was used a fused-silica capillary tube at 280 oC, spray voltage of 5.00 kV, capillary voltage of -35 V, tube lens of -100 V and a 5.0 µL min-1 flow. Full scan analysis was recorded in m/z range from 100-2000. Multiple-stage fragmentations (ESI-MSn) were performed using the collision-induced dissociation (CID) method against helium for ion activation. The first event was a full-scan mass spectrum to acquire data on ions in that m/z range. The second scan event was an MS/MS experiment performed by using a data-dependent scan on the desodiated molecules from the compounds 34

of interest at a collision energy of 30% and an activation time of 30 ms. The product ions were then submitted to further fragmentation in the same conditions, until no more fragments were observed.

3.1.4. UPLC-ESI-IT-MS analysis

UPLC-ESI-IT-MS analysis were carried out in a Thermo Scientific® ultra- performance liquid chromatography equipment, consisting of an Accela AS autosampler, a quaternary Accela pump 600 coupled with the LTQ XL mass spectrometer described above, operating under the same conditions. Chromatographic separations were performed on a non- polar column (Kinetex® core-shell, C18, 1.7 µm, 100 x 2.1 mm, Phenomenex, USA) at room temperature. The mobile phases consisted of eluent A (0.1% formic acid in water, v/v) and eluent B (0.1% formic acid in methanol, v/v). These eluents were delivered at a flow rate of 0.2 mL min-1 with a linear gradient program as follows: 40–100% B from 0 to 5.0 min. After maintaining 100% B for 5 min, the column was returned to its initial condition. Aliquots of 10 µL of the samples were injected into the UPLC-ESI-IT-MS system for analyses using the autosampler. In the UPLC-ESI-IT-MSn experiments for each parent ion of m/z 1405, m/z 1389, m/z 1239 and m/z 1227 we used the four product ions obtained after MS2, MS3 and MS4 experiments and CID of 30 eV against each ion, using the same chromatographic conditions.

3.2. Marine bacteria

3.2.1. Cultivable bacterial communities of marine sediment and invertebrates

3.2.1.1. Sample collection

The marine bacteria studied was isolated from the gastropod Olivancillaria urceus and the starfish Luidia senegalensis collected of the coast of Ubatuba – state of São Paulo, Brazil, as previously described. The sediment sample was collected at the end of the trawl using a van Veen grab sampler deployed from the fishing vessel. The sediment sample was kept wet with sea water and stored in on ice until arrival in the laboratory. In the laboratory, it was kept refrigerated at 4 ºC until processing.

3.2.1.2. Methods for the isolation of bacteria

For the isolation of bacteria, four different pretreatments specific to each type of sample were performed. When seawater was needed, seawater collected from the northern coast of Prince Edward Island, Canada, was filtered over a 0.45 µm cellulose nitrate membrane and autoclaved. Serial dilutions were prepared using sterile sea water. Isolation plates were incubated at 28 ºC for a period of 10 days. (1) Dry-stamp method (dry sediment) (MINCER et al., 2002). A portion (5 g) of wet sediment was dried aseptically in a sterile petri dish (24 h, 37 ºC) and ground lightly with a sterile mortar and pestle. An autoclaved foam plug (diameter 10 mm) was pressed into the ground sediment and excess sediment was dislodged from the plug by gently tapping the edge of the plug. Agar plates were inoculated by stamping the foam plug on the surface of the plate in a circular fashion resulting in a serial dilution effect. Three plates each of media 1-5 were inoculated in this fashion. (2) Dispersial and differential centrifugation technique (DDC - wet sediment) (HOPKINS; MACNAUGHTON; O’DONNELL, 1991). 5.0 g of wet sediment were ground with a sterile mortar and pestle in 10.0 mL of 0.1% (w/v) sodium cholate solution. The ground sediment was aseptically transferred to a 50 mL centrifuge tube and an additional 10 mL of 0.1% (w/v) sodium cholate solution was added along with 30 sterile glass beads (4 mm diameter). The tube was shaken on its side at 60 rpm at 5 °C for 2 h. The mixture was centrifuged at 500g for 2 min and the supernatant saved (DDC1). The precipitate was suspended in 10 mL of 50 mM Tris-HCl buffer (pH 8.0) and stirred for 1 hour at 5 oC. The mixture was then centrifuged at 500g for 1 min. The supernatant was combined with DDC1. The pellet was suspended in 20 mL sodium cholate solution and sonicated in a Misonix Sonicator 3000 (New York, USA) at low power for 1 min. Following sonication 10 mL of sodium cholate solution was added and the tube was shaken on its side at 60 rpm for 1 h at 5 °C. The mixture was centrifuged at 500g for 1 min and the supernatant (DDC2) saved. The pellet was suspended in 10 mL of 50 mM Tris-HCl buffer (pH 8.0) and shaken on its side at 60 rpm for 1 h at 5 oC. The mixture was centrifuged at 500g for 1 min and the supernatant combined with DDC2. Cells present in DDC1 and DDC 2 were collected by centrifugation (12,000g, 4 °C), and suspended in 10 mL of 50 mM Tris-HCl buffer. For media 1-5 0.1 mL of undiluted and a 10-1 dilution of DDC1 and DDC2 were plated on five plates of each medium. For Marine Agar (MA), DDC1 and DDC2 were serially diluted (10-1 to 10-7) and 0.1 mL of each dilution was spread on three plates per dilution. 36

(3) Heat shock method (invertebrates) (MINCER et al., 2002). Invertebrates were thawed at room temperature and in the case of O. urceus shells were aseptically removed once thawed. From the starfish, two arms and the center disc of one specimen were homogenized. From O. urceus, two entire specimens were homogenized. Both invertebrates were mixed with 9 mL of sterile seawater and blended in a stainless steel autoclaved waring blender. The mixture was heated for 60 min at 55 °C to reduce the viability of asporogenous bacteria and then vortexed for 1 minute. The homogenates were serially diluted (10-1 to 10-8) in sterile seawater and 0.1 mL of each dilution was plated on agar plates. For media 1-5, the 10-1 to 10-3 dilutions were plated on each medium in triplicate. For MA, 0.1 mL of each dilution (10-1 to 10-8) was plated in triplicate. (4) Phenol method (invertebrates) (HAYAKAWA; YOSHIDA; IIMURA, 2004). Phenol (1.5% v/v) was added to 10 mL of each invertebrate homogenate and shaken at 200 rpm for 30 minutes. Phenol-treated homogenates were diluted and plated as described for method 3. Bacteria were also isolated from invertebrate homogenates without pretreatment. For these samples a portion (0.1 mL) of serially diluted homogenates (10-1 to 10-8) were plated on triplicate plates of MA.

3.2.1.3. Culture media for bacterial isolation

Five media were used, which had previously been developed for the isolation of Actinobacteria. The compositions of the media are listed below. Culture Medium 1 (HAYAKAWA; NONOMURA, 1987) (per L): humic acid, 1.0 g

(solubilized in 10.0 mL NaOH 0.2 N); Na2HPO4, 0.5 g; KCl, 1.71 g; MgSO4•7H2O, 0.05 g;

FeSO4•7H2O, 0.01 g; CaCO3, 0.02 g; vitamins (thiamine-HCl, riboflavin, niacin, pyridoxine- HCl, inositol, Ca-pantothenate and p-aminobenzoic acid, 0.5 mg each, and 0.25 mg of biotin); agar, 18.0 g; seawater, to 1.0 L; pH 8.0; Vitamins were filter sterilized (0.22 µm) and added to the medium after autoclaving. Culture Medium 2 (ROWBOTHAM; CROSS, 1977)

(per L): KH2PO4, 0.466 g; Na2HPO4, 0.732 g; KNO3, 0.10 g; MgSO4•7H2O, 0.10 g; CaCO3,

0.02 g; sodium propionate, 0.20 g; FeSO4•7H2O, 200 µg; ZnSO4•7H2O, 180 µg; MnSO4•H2O,

20 µg; thiamine•HCl 4 mg; agar, 18.0 g; seawater, to 1.0 L; pH 8.0. Thiamine was filter sterilized (0.22 µm) and added to the medium after autoclaving. Culture Medium 3 (ATHALYE; LACEY; GOODFELLOW, 1981) (per L): yeast extract, 10.0 g; glucose, 10.0 g; agar, 15.0 g; seawater, to 1.0 L; pH 8.0. Culture Medium 4 (VICENTE et al., 2013): 37

mannitol, 0.5 g; peptone, 0.1 g; agar, 18.0 g; seawater, to 1.0 L; novobiocin (20 mg) was filter sterilized and added post sterilization. Culture Medium 5 (VICENTE et al., 2013) (per L): colloidal chitin, 0.5 g; agar, 18.0 g; seawater to 1.0 L. Marine Agar (DifcoTM) (MA) was prepared with deionized water according to the manufacturer’s recommendations. Cycloheximide (50 mg L-1) was added to media 1, 2, 4, and 5 to suppress fungal growth (WILLIAMS; DAVIES, 1965). Rifampicin (5 mg L-1) and streptomycin (15 mg L-1) were added to medium 3 to suppress the growth of fast-growing Streptomyces and to favor the isolation of less frequently isolated Actinobacteria such as Actinoplanes (DWORKIN et al., 2006). Novobiocin (20 mg L-1) was added to media 4 and 5 to suppress the growth of non- actinomycetes (QIU; RUAN; HUANG, 2008). Emerging colonies were purified by serial subculturing on the same medium from which they were first observed. Following initial purification isolates were transferred to MA culture medium and those exhibiting similar morphological characteristics from the same source were grouped and one of each group was selected for identification.

3.2.1.4. Bacteria identification

Bacteria were identified by sequencing of the small subunit (16S) ribosomal RNA gene. To generate template DNA for PCR amplification one colony of each axenic strain was suspended in 50 µL PCR grade DMSO (Sigma). PCR amplification of the 16S rRNA gene was conducted in 50 µL volumes and consisted of the following: EconoTaq® PLUS GREEN 2X Master Mix (25 µL) (Lucigen, Middleton, WI, USA), 0.5 µM of the primers pA (5’- AGAGTTTGATCMTGGCTCAG) and pH (5’-AAGGAGGTGWTCCARCC) and genomic DNA (2.5 µL of template in DMSO) (EDWARDS et al., 1989). Thermal cycling parameters consisted of initial denaturation at 94 °C for 2 min, 30 cycles of 94 °C for 30 s, 60 °C for 30 s and 72 °C for 1.5 min followed by a final extension at 72 °C for 5 min. A negative control, which lacked template DNA (DMSO only), was included in each set of PCR reactions. Amplification was evaluated by agarose gel electrophoresis. Partial sequencing of 16S rDNA amplicons was performed by Eurofins MWG Operon (Huntsville, AL, USA) using the 16S936R primer (5′-GGGGTTATGCCTGAGCAGTTTG) (DUNCAN et al., 2014).

3.2.1.5. Phylogenetic analysis of cultured bacteria

Sequences were analyzed, edited and grouped into operational taxonomic units (OTUs) using the Contig Express application within the Vector NTI version 10.3 software package (Invitrogen, Carlsbad, CA, USA). The level of 16S rDNA sequence identity which corresponds to genomic species demarcation based on a genome average nucleotide identity of 95-96% was recently determined to be 98.65% (KIM et al., 2014). This determination was based on nearly full-length 16S rDNA sequences. As we obtained partial sequences approximately 600-800 bp in length, a more conservative species level 16S rDNA identity cut-off of 99% identity was used to define OTUs. To identify the closest relatives, sequences were compared to those in the NCBI database (http://www.ncbi.nlm.nih.gov/) using the Basic Local Alignment Search Tool (BLAST) (ALTSCHUL et al., 1990). Sequences were aligned using BioEdit version 7.2.5 with the ClustalW tool and phylogenetic analysis of partial 16S rDNA sequences was conducted using the neighbor-joining algorithm (SAITOU; NEI, 1987) based on distances estimated by Kimura's two-parameter model using Molecular Evolutionary Genetics Analysis - MEGA Version 6.0 (TAMURA et al., 2013). Neighbour-joining (NJ) trees were prepared using default settings with complete deletion (FELSENSTEIN, 1985). The robustness of the resulting phylogeny was evaluated by bootstrap analysis of NJ data based on 1000 re-samplings (FELSENSTEIN, 1985).

3.2.1.6. Bacteria cultivation and extraction

Marine Broth (DifcoTM) medium was prepared with Milli-Q water according to the manufacturer’s recommendations and dispensed (7 mL) into culture tubes (150 × 25 mm) containing 3 glass beads (4 mm dia) and sterilized by autoclaving (121 °C for 30 min). Seed cultures were prepared by inoculating an axenic colony into a fresh tube and culturing for three days at room temperature (22-25 °C) and 200 rpm. After this period, 1.0 mL of the broth was transferred to a new tube containing the same medium, which was fermented for one more day under the same conditions. A portion (210 µL) of the second stage seed culture was used to inoculate duplicated culture tubes containing 7 mL of Marine Broth. Tubes were incubated at room temperature and 200 rpm for 7 days. Fermentations were extracted twice with 10 mL ethyl acetate by rapid agitation (200 rpm) at room temperature for one hour. Ethyl acetate extracts were washed with Milli-Q water to remove salts, and then dried under air.

3.2.1.7. Antimicrobial assays

Antimicrobial assays were performed by Martin Lanteigne from the University of Prince Edward Island – UPEI. All microbroth dilution antimicrobial screening was carried out in 96-well plates in accordance with Clinical Laboratory Standards Institute testing standards (NCCLS, 2003) using the following pathogens: methicillin-resistant Staphylococcus aureus ATCC 33591 (MRSA), S. warneri ATCC 17917, vancomycin-resistant Enterococcus faecium EF379 (VRE), Pseudomonas aeruginosa ATCC 14210, Proteus vulgaris ATCC 12454, and Candida albicans ATCC 14035. Extracts were dissolved in sterile 20% DMSO (aq) and assayed at 100 µg mL-1 with a final DMSO concentration of 2% DMSO (v/v; aq). Each plate contained eight uninoculated positive controls (media + 2% DMSO (aq)), eight untreated negative controls (Media + 2% DMSO(aq) + organism), and one column containing a concentration range of a control antibiotic (vancomycin for MRSA, and S. warneri, rifampicin for VRE, gentamicin for P. aeruginosa, ciprofloxacin for P. vulgaris, or nystatin for C. albicans). The optical density (OD) of the plate was recorded using a Thermo Scientific Varioskan Flash plate reader at 600 nm at time zero and then again after incubation of the plates for 22 h at 37 °C. After subtracting the time zero OD600 from the final reading, the percentages of microorganism survival relative to vehicle control wells were calculated.

3.2.1.8. UPLC-HRMS analysis

Extracts that showed antimicrobial activity were analyzed by liquid chromatography– mass spectrometry analysis, carried out with an ESI-HRMS Thermo Scientific® EXACTIVE operating on positive mode with a resolution of 30,000, monitoring a mass range from 200 to 2000 atomic mass units (amu), using a Core-Shell 100 Å C18 column (Phenomenex®

Kinetex, 1.7 µm, 50 × 2.1 mm). A linear solvent gradient from 95% H2O/0.1% formic acid

(solvent A) and 5% CH3CN/0.1% formic acid (solvent B) to 100% solvent B over 4.8 min followed by a hold for 3.2 min with a flow rate of 500 µL/min and 10 µL injection volume was used. The mass spectrometer was hyphenated to an ELSD and UV detector (200 − 600 nm). All analyses were processed using Thermo Xcalibur 2.2 SP1.48 software. To putatively identify ions observed in LC-HRMS analyses, observed molecular weights of pseudomolecular ions were used to search the Antibase 2014 (Wiley-VCH) database. To avoid interference from media components, media blanks were analyzed using the same method and the ions observed subtracted from the sample files. 40

3.2.2. Metabolomic study of the marine actinomycete Verrucosispora sp.

3.2.2.1. Bacteria isolation and identification

Metabolomics study assessed the metabolite production of 3 strains of Verrucosispora sp. isolated from samples collected in the coast of Ubatuba – SP. Strains RKMT_073 and RKMT_111 were isolated from a sediment sample using dry-stamp method (MINCER et al., 2002). Strain RKMT_176 was isolated from a sea star, Luidia senegalensis, using Phenol method (HAYAKAWA; YOSHIDA; IIMURA, 2004). The isolation of RKMT_073 was achieved using Cultivation Medium 2 (ROWBOTHAM; CROSS, 1977) and RKMT_111 and RKMT_176 were isolated using Cultivation Medium 5 (VICENTE et al., 2013), both previously described. All plates were incubated at 30 ºC. When seawater was needed, seawater collected in the northern coast of Prince Edward Island, Canada, was filtered over a 0.45 µm cellulose nitrate membrane and autoclaved. Identification was made by full sequence of the 16S rDNA. Partial sequencing of 16S rDNA amplicons was performed by Eurofins MWG Operon (Huntsville, AL, USA) using the primers 16S936R (5′-GGGGTTATGCCTGAGCAGTTTG), 16S1114F (5′-GCA ACGAGCGCAACCC), 16S530R (5′-GTATTACCGCGGCTGCTGG), 16S27F (5’- AGAGTTTGATCM TGGCTCAG) and 16S1525R (5’-AAGGAGGTGWTCCARCC). Sequences were manually corrected using Chromas Lite 2.1.1 Technelysium Pty Ltd, assembled with online CAP3 Sequence Assembly Program (http://doua.prabi.fr/software/cap3) and the final sequences (~1500bp) compared within the NCBI database sequences (http://www.ncbi.nlm.nih.gov/). The comparison was made by pairwise distance estimation analysis, using Kimura 2 – parameter model in MEGA6 software (TAMURA et al., 2013). All sequences were deposited in GenBank (http://www.ncbi.nlm.nih.gov/genbank/) library (Pending Accession Number).

3.2.2.2. Bacteria fermentation

Three Verrucosispora sp. strains, RKMT_073, RKMT_111 and RKMT_176, were cultivated in nine different types of liquid media each, in duplicate at 30ºC, to evaluate their capacity in the production of secondary metabolites. Thus, the following composition of media was utilized. (1) BFM-1m (LIU; SHEN, 2000) – Dextrin, 20.0 g; Soluble starch, 20.0 g; Beef extract, 10.0 g; Peptone, 5.0 g; (NH4)2SO4, 2.0 g; CaCO3, 2.0 g; Synthetic sea salt 41

(Instant Ocean), 18.0 g and deionized H2O, 1.0 q.s./L. (2) BFM-2m (NAGAOKA et al., 1986) – Soluble starch, 5.0 g; Pharmamedia, 5.0 g; Synthetic sea salt (Instant Ocean), 18.0 g and deionized H2O, 1.0 q.s./L. (3) BFM-3m (GRAZIANI et al., 2005) – MgSO4·7H2O, 0.5 g; KCl,

0.5 g; K2HPO4, 3.0 g; NaCl, 5.0 g, Agar, 0.4g; Glycerol, 12.0 g; Soy peptone, 5.0 g; Synthetic sea salt (Instant Ocean), 18.0 g; Deionized H2O, 1.0 q.s./L; (4) BFM-4m – Nutrisoy, 12.0 g;

NH4Cl, 1.0 g; Dextrose, 12.0 g; Agar, 0.4 g; CaCO3, 1.0 g; NZ-amine A, 3.0 g; Synthetic sea salt (Instant Ocean), 18.0 g and deionized H2O, 1.0 q.s./L. (5) BFM-5m – Pancreatic Digest of

Casein, 17.0 g; Enzymatic Digest of Soybean Meal, 3.0 g; NaCl, 5.0 g; K2HPO4, 2.5 g;

Dextrose, 2.5 g; Synthetic sea salt (Instant Ocean), 18.0 g and deionized H2O, 1.0 q.s./L. (6) BFM-11m (JENSEN et al., 2007) – Starch (potato), 10.0 g; Yeast Extract, 4.0 g; Peptone, 2.0 -1 -1 g; KBr stock (20 g L ), 5.0 mL; FeSO4·7H2O (8 g L , pH 7), 5.0 mL; Synthetic sea salt

(Instant Ocean), 18.0 g and deionized H2O, 1.0 q.s./L. (7) ISP3m – Oat meal, 20.0 g (boiled for 20 min); Trace salts solution (FeSO4·7H2O, 0.1 g, MnCl2·4H2O, 0.1 g, ZnSO4·7H2O in

100 mL of deionized H2O), 1.0 mL; Synthetic sea salt (Instant Ocean), 18.0 g and deionized

H2O, 1.0 q.s./L. (8) ASW-Am (WYCHE et al., 2011) – Soluble starch, 20.0 g; Glucose, 10.0 g; Peptone, 10.0 g; Yeast extract, 5.0 g; CaCO3, 5.0 g; Synthetic sea salt (Instant Ocean), 18.0 TM TM g and deionized H2O, 1.0 q.s./L. (9) Marine Broth (Difco and BBL ) – Peptone, 5.0 g;

Yeast extract, 1.0 g; Iron citrate (III), 0.1 g; NaCl, 19.45 g; MgCl, 8.8 g; Na2SO4, 3.24 g;

CaCl2, 1.8 g; KCl, 0.55 g; NaHCO3, 0.16 g; KBr, 0.08 g; SrCl2, 34.0 mg; H3BO3, 22.0 mg;

Na2SiO3, 4.0 mg; NaF, 2.4 mg; NH4NO3, 1.6 mg; Na2HPO4, 8.0 mg and deionized H2O, 1.0 q.s./L. The seed medium was prepared first by inoculation of a CFU (colony forming unit) of each isolate in a tube containing 7.0 mL of ISP3m liquid medium and fermented for 3 days at 30ºC. Then, 1.0 mL of the fermentation broth was inoculated in a fresh tube containing 7.0 mL of the same liquid medium and fermented for one extra day. Finally, all media for the study (7.0 mL) were inoculated with 210 µL (3% of broth volume) of the fermentation broth and cultivated at 30ºC during 7 days.

3.2.2.3. Extraction and LC-HRMS analysis

All fermentation broths were extracted with two portions of 10.0 mL of ethyl acetate, for 1 hour each time, at 200 rpm. After extraction, all samples were washed with two portions of 5.0 mL of deionized water, the ethyl acetate fractions were dried under air flow and the resulting material diluted to 500 µg mL-1 in methanol. Liquid chromatography–mass 42

spectrometry analyses were carried out with an ESI-HRMS Thermo Scientific® EXACTIVE operating on positive mode as previously described.

3.2.2.4. Preprocessing and statistical analysis

LC−HRMS profiles were analyzed using PCA (principal component analysis) as previously described (FORNER et al., 2013). Briefly, after LC-HRMS analysis, the files were converted from .RAW (proprietary format) to .CDF (non-proprietary format) extension using XCalibur software. All the files were then processed using MZmine 2 (PLUSKAL et al., 2010), where the converted files were submitted to the steps of mass detection, chromatogram building, deisotoping, alignment and exportation. Mass detection step generates a list of exact masses for each scan in the analysis. For this step, it was chosen a noise level value of 1.0x104 counts s-1 by comparison with methanol and media blanks. Peaks presenting a lower intensity were therefore not detected and not included in the mass list for further analysis. The next step, chromatogram building, builds a chromatogram for each mass that can be detected in all scans from the mass list generated in the previous step. Then, deisotoping step was carried out for the selection of the representative ions in isotopic patterns. The last step in MZmine 2 software was the alignment mode, where m/z and retention time (rt) data were combined and all samples aligned in a single file. The alignment mode allows the conversion of a three dimensional dataset to a two dimensional dataset, where buckets are created with the peaks intensity. For the peaks that are not present in the sample, a value of zero was assigned. Finally, the aligned peaks were exported as a .csv (comma separated value) file. Standardization and artifact suppression were carried out in Microsoft Excel 2010. First, a presence-absence standardization was achieved transforming the dataset in a binary pattern, where intensities greater than zero were given a value of one and intensities equal to zero remained zero. For the easy visualization and pattern recognition, values of “1” within the dataset are shown as black, while values of “0” are white. Ions detected in MeOH and media blanks were removed from samples for media and artifact suppression. Ions that were not consistent in both replicates were excluded. Statistical analysis was carried out using The Unscrambler (Camo software). Principal component analysis was performed for comparison and for identification of putatively new compounds. Cluster analysis was performed using Ward’s Method and Squared Euclidean distances, as previously described (FORNER et al., 2013; WARD, 2011). 43

The number of compounds was estimated considering the detected adducts with approximate retention time. When no adduct was detected, the ion was considered as one compound. All adducts corresponding to each compound are shown in Table S1 – Supplementary Material.

3.2.2.5. Identification of compounds

After detection and recognition of adducts the ions were compared in Antibase 2012 Wiley® (LAATSCH, 2012) through search of the [M+H]+ and [M+Na]+. Identification of known metabolites was further confirmed using the Scifinder database (http://www.scifinder.cas.org) and searching by molecular formula (Table S1 – Supplementary Material – Page 107).

3.2.3. Biotransformations in bacterial isolates

3.2.3.1. Bacteria isolation and identification

Strain RKMT_070 was isolated from a sediment sample collected in the coast of Ubatuba – SP, using dry-stamp method (MINCER et al., 2002), in triplicate, in agar plates containing Culture Medium 2 (ROWBOTHAM; CROSS, 1977) and identified by full sequence of the 16S rDNA using the same primers as previously described for the Verrucosispora sp. isolates. Reference and experimental sequences were aligned using BioEdit version 7.2.5 with the ClustalW tool and phylogenetic analysis of full 16S rDNA sequences was conducted using the neighbor-joining algorithm (SAITOU; NEI, 1987) based on distances estimated by Kimura's two-parameter model using Molecular Evolutionary Genetics Analysis - MEGA Version 6.0 (TAMURA et al., 2013). Neighbour-joining (NJ) tree was prepared using default settings with complete deletion (FELSENSTEIN, 1985). The robustness of the resulting phylogeny was evaluated by bootstrap analysis of NJ data based on 1000 re-samplings (FELSENSTEIN, 1985). The comparison was also made by pairwise distance estimation analysis, using Kimura 2 – parameter model in MEGA6 software (TAMURA et al., 2013). Final sequence was deposited in GenBank (http://www.ncbi.nlm.nih.gov/genbank/) library (Pending Accession Number).

3.2.3.2. Fermentation, extraction and LC-HRMS analysis

Strain RKMT_070 was cultivated in Marine Broth (MB – DifcoTM), at room temperature, to evaluate its capacity in the production of secondary metabolites. Thus, the following composition of media was utilized: Marine Broth (DifcoTM and BBLTM) – Peptone,

5.0 g; Yeast extract, 1.0 g; Iron citrate (III), 0.1g; NaCl, 19.45 g; MgCl, 8.8 g; Na2SO4, 3.24 g; CaCl2, 1.8 g; KCl, 0.55 g; NaHCO3, 0.16 g; KBr, 0.08 g; SrCl2, 34.0 mg; H3BO3, 22.0 mg;

Na2SiO3, 4.0 mg; NaF, 2.4 mg; NH4NO3, 1.6 mg; Na2HPO4, 8.0 mg and deionized H2O, 1.0 q.s./L. The seed medium was prepared first by inoculation of a CFU (colony forming unit) of the isolate in a tube containing 7.0 mL of MB liquid medium and fermented for 3 days at room temperature. Then, 1.0 mL of the fermentation broth was inoculated in six fresh tubes containing 7.0 mL of the same liquid medium and fermented for one extra day. Finally, 120 tubes containing 7.0 mL of MB were inoculated with 210 µL (3% of broth volume) of the fermentation broth and cultivated at room temperature during 7 days at 200 rpm. Moreover two 2.0 L capacity baffled flasks containing 250 mL of MB were inoculated with 7.5 mL of the fermentation broth and cultivated in the same conditions. Tubes containing the fermentation broths were combined in a separation funnel and extracted with two portions of 500 mL of ethyl acetate. After extraction, each ethyl acetate phase was washed with two portions of 300 mL of deionized water, combined and dried. The same procedure was carried out with the fermentation broth from cultivation in baffled flasks. The remaining fermentation broth from both tubes and baffled flasks were separately extracted with HP-20 resin in a proportion of 50g/L of broth. The HP-20 was, therefore, extracted consecutively with 400 mL of H2O, MeOH/H2O 1:1 and MeOH for 30 min each and the solvent dried. All samples (ethyl acetate and HP-20 fractions from tubes and baffled flasks) were analyzed by LC-HRMS in a concentration of 500 µg mL-1 in methanol for comparison. Liquid chromatography–mass spectrometry analyses were carried out with an ESI- HRMS Thermo Scientific® EXACTIVE operating on positive mode as previously described. Compounds detected were compared to Antibase 2012 Wiley® for the identification of possible known metabolites previously reported as produced by bacteria.

3.2.3.3. Isolation and identification of compounds

Fractions MeOH from HP-20 of the fermentation broth from tubes and baffled flasks were combined and the compounds purified using a Sunfire C18 column (5 µm, 250 × 10.00 ® mm, 110 Å, Waters ). Deionized H2O 0.1% formic acid (solvent A) and acetonitrile 0.1% formic acid (solvent B) were used with a flow rate of 3 mL/min and a sample concentration of 10 mg/mL. The substances were separated starting with an isocratic elution using 45% of solvent B over 15 min, followed by a linear gradient increasing from 45% solvent B to 90% solvent B over 5 min, followed by another isocratic method using 90% of solvent B for 5 min, finishing with a decrease of solvent B from 90% to 45% in 2 min and a hold of solvent B 45% over 3 min. Compounds were detected with a Waters® MS detector and twelve fractions were collected. Purity of the compounds was evaluated by LC-HRMS analysis as previously described and identification was made by NMR spectroscopy carried out on a 600 MHz Bruker® Avance III spectrometer equipped with a 1.7 mm inverse probe. Chemical shifts (δ) are reported in ppm and were referenced to the MeOD residual peaks at δH 3.31 ppm and δC 49.00 ppm.

4. Results

There are a number of published data about the marine invertebrates associated with shrimp trawling in the coast of São Paulo State and most of them were compiled in the manuscript of Branco et al. (2015). In our work, we evaluated the number of captured specimens and measured the biomass related to them. A total of 230 kg of biomass was collected in all the six trawling. The amount of commercial shrimp was only 15 kg (6.5%), which lead to the proportion of 1:15 between the target organism and by-catch fauna. In our study, six trawls captured 132 species. In addition to the target shrimps [Litopenaeus schmitti (white shrimp), Farfantepenaeus brasiliensis and Farfantepenaeus paulensis (pink-shrimp) and Xiphopenaeus kroyeri (seabob)], other six species of shrimp with little or no commercial interest were also identified in negligible amounts Artemesia longinaris, Hepomadus sp., Rimapenaeus constrictus, Exhippolysmata oplophoroides, Sicyonia parri and Sicyonia dorsalis), all belonging to the superfamily Penaeoidea. Fishes (135 kg) accounted for 54% of the by-catch fauna and crustaceans (55 kg) for 22%. After these, the most representative phylum was Mollusca (16 kg, 6% biomass; 7 species, 7%), followed by Echinodermata (9 kg, 3.5% biomass; 6 species, 4.5%), Cnidaria (0.5 kg, 2% biomass; 4 species, 3%) and one Sipuncula (0,03 g). From these, some invertebrates could not be analyzed due to problems of maintenance and/or identification (sponges, ascidians and jellyfishes), which amounted to 4,836 specimens in a total of approximately 25 kg biomass (11%).

4.1. Mass spectrometry study of the starfish Luidia senegalensis

In this study, we assessed the chemical profile of the saponins of the starfish Luidia senegalensis collected as by-catch fauna of the shrimp fishery in Brazil, as a result of trawling employed by traditional fishers, who use to discard these organisms. No chemical investigation of this marine invertebrate could be found in the literature. Starfish present many interesting compounds, which are usually derived from sulphated steroids, with a wide variety of substitutions in the aglycone core (E.g. hydroxylation pattern, side chain arquitecture) (DE MARINO et al., 2003; DEMEYER et al., 2014; DONG et al., 2011; MINALE et al., 1985; POPOV et al., 2014). 47

The putative structure of the saponins detected in L. senegalensis were associated closely with those found in Aphelasterias japonica (POPOV et al., 2014), Asterias rubens (DEMEYER et al., 2014), L. maculata (MINALE et al., 1985) and in L. quinaria (DE MARINO et al., 2003), besides those described in the review published by Dong et al.(2011). In view of the fact that many of the starfish saponins contains similar molecular weight but different aglycone design and a number of oligosaccharide chains, the confirmation of the structures requires further chemical investigation using additional spectrometric techniques (e.g Nuclear Magnetic Resonance, Circular Dichroism, etc). Therefore, it is important to reinforce that the main goal of this work was not to fully identify the saponins, but instead to present an overview of the chemical composition of the discarded material considered as by- catch of the shrimp fishery.

4.1.1. SPE fractionation of the crude extract

The matrix used in these experiments is a mixture of inorganic salts (NaCl), organic salts (sulphated saponins), and amphoteric compounds (amino acids). To extract the secondary metabolites from the hydroethanolic crude extract and remove inorganic salts and amino acid impurities, we use a solid phase extraction in a C18 reversed phase (RP18 SPE) cartridge. After sequential elution with water/methanol gradient, thin layer chromatography (TLC) analyses in silica gel plates indicated that inorganic salts and amino acids remained only in the aqueous fraction (revealed as reddish spots when sprayed with ninhydrin reagent), whereas saponins appeared as purple spots in the hydromethanolic fraction (HF) (WAGNER; BLADT, 1996). Besides, UPLC-ESI-IT-MS analysis did not evidenced peaks corresponding to amino acids, which also shows that the SPE procedure used was efficient for the separation of the secondary metabolites from these starfishes, leading to the elimination of the amino acids.

4.1.2. DFI-ESI-IT-MSn analyses

In order to obtain a preliminary fingerprint about the chemical composition of L. senegalensis, HF was directly injected into the ESI source of the ion trap. We have tested different mass spectrometry conditions and decided to use the negative ionization and optimized conditions as presented in the Experimental Section. This fact agrees with the literature, which reports intense [M-Na]- signals in the range of m/z 1100-1400, corresponding 48

+- to ionized saponins containing Na O3SO- groups. Demeyer et al. (DEMEYER et al., 2014) observed that, since the sulphate group is commonly attached to the position 3 of the aglycone core, the most intense signals of the mass spectra contains the aglycone nucleus. In this case, intense signals were observed in the m/z 1200-1500 Da range of the full scan mass spectra, that may be assigned to the presence of negatively ionized saponin ions [M-Na]- (Figure 1).

F: ITMS - c ESI Full ms [310.00-2000.00] 1243.6 100

90 1239.8

50 1389.6 1405.7 40

30 1227.6

20 1143.5 1306.6 1213.9 1257.7 1093.7 1163.4 1289.7 1373.8 10

0 1100 1150 1200 1250 1300 1350 1400 1450 1500 m/z Figure 1. Full scan DFI-ESI-IT-MSn (Negative Ionization) mass spectrum of Luidia senegalensis showing the peaks corresponding to the saponins

4.1.3. UPLC-ESI-IT-MS analysis

Figure 2 displays the analysis of the HF from L. senegalensis by UPLC-ESI-IT-MS. After optimization of the solvent composition and elution gradient, the chromatogram showed a reasonable baseline separation for the peaks, which could be analyzed in a time interval of less than 10 min with little interference. It is worth mentioning that no baseline separation and strong peak tailing were observed when we performed chromatographic runs under the same conditions, but without acidifying the eluent mixtures with 0.1% formic acid.

100 2 4 NL 6 B 50 1 7 3 5 Ls_ 0 100 1 NL m 50 Ls_ 0 100 2 NL m 50 Ls_ 0 100 NL 4 m 50 Ls_ 0 100 NL 3 m 50 5 Ls_ 0 100 6 NL m 50 Ls_ 0 100 7 NL m 50 Ls_ 0 0 1 2 3 4 5 6 7 8 9 Time (min) Figure 2. UPLC-ESI-IT-MS analysis of the saponins present in Luidia senegalensis. Base Peak Ion - BPI (above) and extracted chromatograms of the ions 1: m/z 1405; 2: m/z 1389; 4: m/z 1239; 3 and 5: 1227; 6: m/z 529; 7: m/z 547

Even using very strong source conditions (E.g. -80 eV), UPLC-ESI-IT-MS experiments were not enough to produce extensive fragmentation of the precursor ions contained into the chromatographic peaks. Therefore, we performed UPLC-ESI-IT-MSn experiments using the precursor ion of each chromatographic peak as well as the MS2, MS3 and MS4 product ions observed after the DFI-ESI-IT-MSn experiments. Combination of DFI- ESI-IT-MSn, UPLC-ESI-IT-MS and UPLC-ESI-IT-MSn led to the data displayed in Table 2, which shows the retention times, main fragment ions and the tentative assignment of the compounds. The putative identification is discussed ahead and the chemical structures of the compounds are shown in Figure 3. 50

1 2 ∆24,25

Figure 3. Aglycones of the steroidal saponins detected in Luidia senegalensis. R corresponds to the sugar sequence described in Table 2

Table 2. Proposed saponins present in Luidia senegalensis detected by ESI-IT-MSn (1), UPLC-ESI-IT- MS and UPLC-ESI-IT-MSn (2) in negative ion mode

Tentative assignment Peak R Precursor t Fragment ions (m/z) MW Aglycone (min) íon (m/z) Sugar (no sugar sequence units) 1305(1,2), 1243(2), 1143(1,2), 1097(1,2), Hex-dHex- 1 4.74 1405 997(1,2), 935(1,2), 1428 1 (6) Hex-Pent(- 835(1,2), 789(2), 657(2), dHex)-Hex 557(2) 1289(1,2, 1243(2), 1143(1,2), 1097(1,2), dHex-dHex- 2 5.06 1389 997(1,2), 935(2), 1412 1 (6) Hex-Pent(- 851(1,2), 835(1,2), dHex)-Hex 689(1,2), 395(2) 1127(1,2), 1081(1,2), 981(1,2, 935(1,2), dHex-dHex- 3 5.34 1227 919(1,2), 849(2), 1250 1 (5) Pent(- 835(1,2), 773(1,2), dHex)-Hex 641(1,2), 495(1,2) dHex-dHex- 1093(2), 947(2), 801(2), 4 5.70 1239 1262 2 (5) dHex(- 655(2), 493(2) dHex)-dHex 1127(1,2), 1081(1,2), 981(1,2, 935(1,2), dHex-Hex- 5 5.93 1227 919(1,2), 849(2), 1250 1 (5) Pent(- 835(1,2), 773(1,2), dHex)-dHex 641(1,2), 495(1,2) 6 6.24 529 ND - - - 7 7.65 547 ND - - - ND: Not determined; dHex: deoxyhexose; Hex: hexose; Pent: Pentose.

In brief, the ions of [M-Na]- were, in most cases, the major ones, and their main fragmentation pathway were neutral loss of sugar units. Typical losses of 132 Da, 146 Da and 51

162 Da were assigned to the presence of pentose [Pent, e.g., xylose (132)], deoxyhexose [dHex, E.g. fucose or quinovose (146)] and hexose [Hex, E.g. glucose, galactose or DXHU (162 Da)]. In some cases, we observed the simultaneous loss of two sugar units. Side chain fragmentation with 23-oxo substitution led to losses of 100 Da, due to the low energy McLaffery rearrangement of 6-member transition states, which generates the neutral molecule

C6H12O (4-methylpent-1-en-2-ol) (DATTA; TALAPATRA; SWARNAKAR, 2015; DONG et al., 2011). These typical fragment ions associated with sugar and/or side chain losses were also observed in the MSn spectrum of many sulphated saponins from starfish previously described the literature (DATTA; TALAPATRA; SWARNAKAR, 2015; DEMEYER et al., 2014; DONG et al., 2011; POPOV et al., 2014). Using direct injection and CID dissociation, MS2 fragmentation of the precursor ion at m/z 1405 [M-Na]- led to product ion at m/z 1305 [M-100-Na]-, assigned to the loss of the 3 neutral molecule C6H12O, due to the McLafferty rearrangment. MS fragmentation of the parent ion at m/z 1305 gave the product ion at m/z 1143 [M-100-162-Na]- (loss of a hexose unit). MS4 fragmentation of the precursor ion of m/z 1143 led to the product ion at m/z 997 [M-100-162-146-Na]- (loss of a deoxyhexose). MS5 fragmentation of the precursor ion at m/z 997 led to the product ions at m/z 851 [M-100-162-146-162-Na]- (loss of a hexose) and at m/z 835 [M-100-162-146-146-Na]- (loss of a deoxyhexose), thus evidencing the presence of a branched saccharide chain. Due to the low abundance, further MSn experiments were not practicable. Figure 4 displays an example of the UPLC-ESI-IT-MSn analysis performed with the ion of m/z 1405 and the product ions up to MS4 (obtained from the DFI-ESI-IT-MSn experiments), present in the peak with retention time 4.74 min. We could observe the production of several product ions, which corresponds to the fragmentation sequence 1405 → 1243 (loss of 162, hexose) → 1097 (loss of 146, deoxyhexose) → 935 (loss of 162, hexose) → 789 (loss of 146, deoxyhexose) → 657 (loss of 132, pentose) → 557 (loss of 100,

C6H12O). Despite we could not observe the loss of the hexose attached to the aglycone, the combination of these data was consistent with the molecular formula C62H101O33SNa and allowed us to suggest the presence of the aglycone 1 attached to the sugar sequence Hex- dHex-Hex-Pent(-dHex)-Hex. 52

Figure 4. Mass spectrum obtained after UPLC-ESI-IT-MSn experiment using the precursor ion

MS2 fragmentation of the precursor ion at m/z 1389 [M-Na]- led to product ion at m/z 1289 [M-100-Na]-. MS3 fragmentation of the parent ion at m/z 1289 gave the product ion at m/z 1143 [M-100-146-Na]- (loss of a deoxyhexose unit). MS4 fragmentation of the precursor ion of m/z 1143 led to the product ion at m/z 997 [M-100-146-146-Na]- (loss of a deoxyhexose). MS5 fragmentation of the precursor ion at m/z 997 led to the product ions at m/z 851 [M-100-146-146-146-Na]- (loss of a deoxyhexose) and at m/z 835 [M-100-146-146- 162-Na]- (loss of a hexose), again showing the presence of a branched saccharide chain. MS6 experiment of the parent ion at m/z 835 led to the product ion at m/z 689 [M-100-146-146- 162-146-Na]-. The UPLC-ESI-IT-MSn analysis of the peak with retention time 5.06 min produced the fragmentation sequence 1389 → 1243 (loss of 146, deoxyhexose) → 1097 (loss of 146, deoxyhexose) → 935 (loss of 162, hexose). These data are in agreement with the molecular formula C62H101O32SNa, which suggest the presence of aglycone 1 attached to the sugar sequence dHex-dHex-Hex-Pent(-dHex)-Hex. MS2 fragmentation of the precursor ion at m/z 1227 [M-Na]- led to product ions at m/z 1127 [M-100-Na]- (loss of the side chain due to McLafferty rearrangement) and at m/z 1081 [M-146-Na]- (loss of the terminal sugar moiety). Such co-occurrence of different fragmentation routes was already described in the literature and can be assigned to the presence of saponin isomers (DATTA; TALAPATRA; SWARNAKAR, 2015; DONG et al., 2011). Therefore, we first choose to proceed MS3 experiments with the fragmentation of the parent ion at m/z 1127, which gave the product ion at m/z 981 [M-100-146-Na]- (loss of a 53

deoxyhexose) and at m/z 1109 [M-100-18-Na]- (loss of water). Demeyer et al. (2014) demonstrated that the water loss is associated with the presence of a DXHU residue in the oligosaccharide chain. MS4 fragmentation of the precursor ion at m/z 981 gave the product ion at m/z 835 [M-100-146-146-Na]-. Due to the low abundance of the m/z 835 ion we could not perform further experiments. This fragmentation pattern and the molecular weight suggest molecular formula C56H91O27SNa, which prompted us to propose the presence of the aglycone 1 attached to the sugar sequence dHex-dHex-Pent(-dHex)-Hex. On the other hand, MS3 fragmentation of the precursor ion at m/z 1081 led to the product ions at m/z 935 [M-146-146- Na]- (loss of a second deoxyhexose unit) and at m/z 919 [M-146-162-Na]- (loss of an hexose), thus suggesting the presence of a branched saccharidic chain at C6. MS4 fragmentation using the parent ion at m/z 919 led to the product ion at m/z 773 [M-146-162-146-Na]- (loss of a second deoxyhexose unit), at m/z 641 [M-146-162-146-132-Na]- (loss of a pentose unit) and at m/z 495 [M-146-162-146-132-146-Na]- (loss of a deoxyhexose). MS4 fragmentation of the ion at m/z 935 led to the same product ion at m/z 773 [M-146-146-162-Na]- (loss of the hexose moiety). MS5 fragmentation of the precursor ion at m/z 773 led to the product ion at m/z 641 [M-146-146-162-132-Na]- due to the loss of a pentose unit. MS6 fragmentation of the precursor ion at m/z 641 led to the product ion at m/z 495 [M-146-146-162-132-146-Na]-, consistent with the loss of a final deoxyhexose moiety. All these data suggest molecular formula C56H91O27SNa, and are compatible a saccharide sequence dHex-Hex-Pent(-dHex)- dHex linked to the aglycone 1. These two molecules could account to the presence of two chromatographic peaks for the extracted ion of m/z 1227, at 5.34 and 5.93 min. MSn fragmentation of the ions contained in these two chromatographic peaks did not led to good fragmentation, probably due to the low abundance, and were not considered. MS2 fragmentation of the precursor ion at m/z 1239 [M-Na]- led to product ion at m/z 1093 [M-146-Na]-, corresponding to the loss of a deoxyhexose unit. No fragmentation of the side chain was observed. Further MSn fragmentation led to the sequential losses of four deoxyhexose moieties, producing the ions at m/z 947 [M-146-146-Na]-, at m/z 801 [M-146- 146-146-Na]-, at m/z 655 [M-146-146-146-146-Na]-. In addition, fragmentation of the parent ion at m/z 655 led to the product ion at m/z 493 [M-146-146-146-146-162-Na]-, assigned to the loss of one hexose. The UPLC-ESI-IT-MSn analysis of the peak with retention time 5.70 min produced the fragmentation sequence 1239 → 1093 (loss of 146, deoxyhexose) → 947 (loss of 146, deoxyhexose) → 801 (loss of 146, deoxyhexose) →655 (loss of 146, deoxyhexose) → 493 (loss of 162, hexose). All these data are compatible with an asterosaponin with the formula C55H87O27SNa. Since the aglycone differs in 2 Da from 54

aglycone 1, we propose the presence of an additional double bond, like in compound 2, attached to the sugar sequence dHex-dHex-dHex(-dHex)-Hex-Agycone. This proposal is similar to Ruberoside G, which was reported in the sea star Asterias rubens (DONG et al., 2011). Peaks 6 and 7 did not present fragmentation in the UPLC-ESI-IT-MS experiments, and they might be due to the presence of polyhydroxysteroids, of large occurrence in starfish (CLARCK; DOWNEY, 1992). No evidence for either 20-hydroxy or 24-methyl side chain (loss of 114 Da due to a McLafferty rearrangment) was detected in our ESI-IT-MSn experiments. Regarding UPLC analyses, separation of peaks containing structurally similar complex compounds bearing the same aglycone but different saccharide chain arrangements was easily achieved with a very short run time (E.g. hexasaccharide saponins: m/z 1405/1389; pentasaccharide saponins: m/z 1227).

4.2. Marine bacteria

4.2.1. Cultivable bacterial communities of marine sediment and invertebrates

4.2.1.1. Bacteria isolation and identification

A total of 230 isolates were picked from initial isolation plates inoculated with dilutions of the two invertebrate and single sediment sample. Initial dereplication of strains obtained from each sample was conducted on the basis of morphological characteristics (e.g. colony shape, size, and color) and resulted in the selection of 134 strains for identification by partial 16S rRNA gene sequencing. The distribution of isolates between samples was as follows: sediment – 50 isolates, L. senegalensis – 53 isolates, O. urceus – 31 isolates. In total, 64 OTUs were obtained using a 99% sequence identity cutoff. Isolates were distributed among 28 genera belonging to the phyla Proteobacteria (8 OTUs), Actinobacteria (22 OTUs) and Firmicutes (34 OTUs) (Tables 3 and 4). The wet sediment yielded the fewest isolates (9) and lowest diversity of genera with only Pseudoalteromonas (89%; 1 OTU) and Vibrio (11%; 1 OTU) recovered from this sample (Tables 3 and 4). A substantially higher number of isolates, OTUs (41isolates/18 OTUs) and a greater diversity of genera were obtained from the dried sediment (Tables 3 and 4). The majority of isolates were Firmicutes (53% of isolates/7 OTUs); Bacillus spp. 55

constituted the majority of this group (81%, 6 OTUs). Actinobacteria were the next most abundant group (40% of isolates/8 OTUs). Five actinobacterial genera were obtained, with Micromonospora sp. accounting for 66% of isolates (5 OTUs). Other Actinobacteria obtained from dried sediment, which are notable due to their propensity to produce bioactive natural products, was a single Streptomyces isolate and two Verrucosispora isolates. The remaining isolates (7%) were Proteobacteria belonging to the genera Wenxinia and Erythrobacter (1 OTU each). Among the marine invertebrates, the sea star L. senegalensis yielded the largest number of isolates and the greatest taxonomic diversity, with 53 isolates and 28 OTUs distributed among 20 genera (Tables 3 and 4). Firmicutes and Actinobacteria accounted for the overwhelming majority of isolates from L. senegalensis (50% and 46% of isolates, respectively), while Proteobacteria (Pseudomonas and Psychrobacter) accounted for the remaining 4%. Among the Firmicutes, Bacillus spp. were the most abundant (29%) followed by seven other genera each accounting for 2-9% of isolates. The proportions of Actinobacteria isolates were more evenly distributed among 10 genera: Micromonospora (9%), Kocuria (9%), Streptomyces (6%), Rhodococuccus (6%), Cellulosimicrobium (4%), Dietzia (4%), Arthrobacter (2%), Microbacterium (2%) and Serinicoccus (2%). The snail O. urceus yielded 31 bacteria isolates belonging to 16 OTUS and 10 different genera (Tables 3 and 4). Firmicutes accounted for 61% of isolates, with Bacillus spp. constituting the majority (39%). Two Proteobacteria genera were recovered from O. urceus (Psychrobacter spp. - 18%, Pseudoalteromonas spp. - 3%), while the Actinobacteria isolates could be assigned to four genera (Kocuria sp.- 3%, Serinicoccus sp. - 6%, Kytococcus sp. - 6%, Micrococcus sp. - 3%). Among all isolation media, Media 4 and 5 were the most selective for Actinobacteria, as the greatest number of actinobacterial isolates (17) and genera (3) were recovered on these media (Tables 3 and 4). The greater selectivity for Actinobacteria corresponded to a lower recovery rate of Proteobacteria and Firmicutes on these media. Medium 3 yielded the fewest isolates (3), all of which were Firmicutes. The greatest number of isolates was obtained from Media 1 (20) and 2 (38), however, these media were less selective for Actinobacteria, as Firmicutes were primarily isolated from these media (41 isolates). MA also yielded a high number (45) of taxonomically diverse isolates and remarkably Actinobacteria were more abundant (21 isolates) than Firmicutes and Proteobacteria. Evaluating the pretreatment methods utilized on the sediment sample, dispersal and differential centrifugation technique (DDC) was less productive than the dry stamp method as it yielded only nine Proteobacteria strains belonging to two genera. The dry-stamp method was a better approach for the study of sediments, since it yielded a higher number of total 56

isolates (41) and a greater variety of genera (10). In these experiments the dry-stamp method was superior to the DDC technique for the isolation of Actinobacteria, as indicated by the isolation of Streptomyces (1 isolate), Micromonospora (24 isolates) and Verrucosispora (2 isolates) strains, which are a prolific source of bioactive natural products as discussed previously. Among the methods used to isolate bacteria from the invertebrate samples, the heat-shock method provided a higher total number of isolates (30) and greatest variety of genera (14). The heat shock approach also resulted in the isolation of the greatest number of Actinobacteria isolates (10), which were identified as Micromonospora spp. (4), Kytococcus spp. (1), Cellulosimicrobium spp. (2), Micrococcus spp. (1) and Rhodococcus spp. (2). The phenol method resulted in the preferential isolation of Firmicutes (14 isolates) compared to Actinobacteria (2 isolates) and Proteobacteria (0 isolates). Plating the untreated invertebrate homogenate on MA resulted in the isolation of the greatest number of Actinobacteria strains (18) with the greatest taxonomic diversity (8 genera) (Tables 3 and 4). Other than a single Streptomyces sp., this non-selective approach yielded amycelial taxa belonging to the orders Micrococcales and Corynebacteriales. Not surprisingly, this approach also yielded the highest numbers of Firmicutes and Proteobacteria. Overall, heat-shock and phenol pretreatments selected for the isolation of gram-positive bacteria (Firmicutes and Actinobacteria), as no Proteobacteria (gram-negative) were obtained using these pretreatments (Table 3).

Table 3. Summary of isolates obtained from different samples using different pretreatments and isolation media. Numbers in represent “number of isolates/number of genera”

Sample Dry Sediment (41/10) Wet Sediment (9/2) Snail (31/12) Sea star (53/20) Culture Media 1 2 3 4 5 1 2 3 4 5 MA 1 2 3 4 5 MA 1 2 3 4 5 MA Actinobacteria

Untreated – – – – – – – – – – – – – – – – 4/3 – – – – – 14/7 DDC – – – – – 0/0 0/0 0/0 0/0 0/0 0/0 – – – – – – – – – – – – Dry-stamp 0/0 5/5 0/0 1/1 10/2 – – – – – – – – – – – – – – – – – – Heat-shock – – – – – – – – – – – 1/1 1/1 0/0 0/0 1/1 – 1/1 2/2 0/0 3/1 2/2 – Phenol – – – – – – – – – – – 0/0 0/0 0/0 0/0 0/0 – 0/0 0/0 0/0 1/1 1/1 – Proteobacteria

Untreated – – – – – – – – – – – – – – – – 6/2 – – – – – 8/4 DDC – – – – – 0/0 0/0 0/0 2/1 1/1 6/2 – – – – – – – – – – – – Dry-stamp 2/2 1/1 0/0 0/0 0/0 – – – – – – – – – – – – – – – – – – Heat-shock – – – – – – – – – – – 0/0 0/0 0/0 0/0 0/0 – 0/0 0/0 0/0 0/0 0/0 – Phenol – – – – – – – – – – – 0/0 0/0 0/0 0/0 0/0 – 0/0 0/0 0/0 0/0 0/0 – Firmicutes

Untreated – – – – – – – – – – – – – – – – 5/3 – – – – – 2/2 DDC – – – – – 0/0 0/0 0/0 0/0 0/0 0/0 – – – – – – – – – – – – Dry-stamp 6/2 15/5 0/0 1/1 0/0 – – – – – – – – – – – – – – – – – – Heat-shock – – – – – – – – – – – 4/2 5/3 1/1 0/0 0/0 – 5/3 1/1 2/1 1/1 2/2 – Phenol – – – – – – – – – – – 1/1 2/1 0/0 0/0 0/0 – 4/3 6/3 0/0 0/0 1/1 – 58

4.2.1.2. Phylogenetic analysis

All strains cultivated from the sediment sample and the two invertebrates were closely related to previously cultivated bacteria. In all cases the 16S rRNA gene sequences of the isolated bacteria shared >99% sequence similarity with cultured type strains (Table 4). The phylogenetic relationships of bacteria isolated in this study and selected type strains are shown in Figures 5-7. In general, each of the three samples examined in this study yielded a taxonomically distinct set of bacterial isolates, as evidenced by the small number of OTUs (8) formed by isolates obtained from more than one sample (Figures 5, 6, and 7; isolates with more than one symbol following the strain number of the OTU representative).

Table 4. Taxonomic affiliation of representative OTUs of bacteria isolated from Brazilian sediment and the invertebrates L. senegalensis and O. urceus. Type strains for comparisons were identified from BlastN searches of the GenBank 16S rRNA gene sequence database. Source: O - Snail; L - Sea Star; D - Dry Sediment; W - Wet Sediment. Pretreatment: U - Untreated; DS - Dry-stamp; H - Heat; P - Phenol; DDC2 - Dispersial and Differential Centrifugation Representative Closest Related Type Strain Kimura 2- Sourceb Pretreatmentc Mediumd Accession No Lenght (bp) isolate (nº of isolatesa) (GenBANK Accession No) p distance Actinobacteria (22d) RKMT_021 (2) O U MA KU501352 822 Kytococcus sedentarius DSM 20547 (X87755.1) 100 RKMT_022 (2) O U MA KU501350 828 Serinicoccus marinus JC1078 (AY382898.2) 99.83 RKMT_027 (3) L U MA KU501343 825 Streptomyces albidoflavus NBRC 13010 (AB184255.1) 100 RKMT_028b (2) L U MA KU501351 822 Dietzia cinnamea DSM 44904 (FJ468339.3) 100 RKMT_034 (1) L U MA KU501353 834 Serinicoccus profundi 0714S6-1 (EU603762.1) 99.3 RKMT_039 (1) L U MA KU501361 721 Arthrobacter sanguinis CCUG 46407 (EU086805.1) 100 RKMT_042 (6) L; O U MA KU501344 829 Kocuria rhizophila TA68 (Y16264.1) 100 RKMT_053 (1) L U MA KU501354 813 Microbacterium esteraromaticum DSM 8609 (Y17231.1) 100 RKMT_071 (1) D DS 2 KU501362 765 Streptomyces carpaticus NRRL B-16359 (DQ442494.1) 100 RKMT_075 (1) D DS 2 KU501355 762 Micromonospora avicenniae 268506 (JQ867183.1) 100 RKMT_091b (1) D DS 2 KU501356 825 Dermacoccus barathri MT2.1 (AY894328.1) 100 RKMT_093 (1) D DS 2 KU501357 828 Kocuria atrinae P30 (FJ607311.1) 100 RKMT_100 (1) D DS 5 KU501363 798 Micromonospora aurantiaca ATCC 27029 (NR_074415.1) 100 RKMT_102 (1) D DS 5 KU501364 760 Micromonospora auratinigra TT1-11 (AB159779.1) 100 RKMT_110 (1) D DS 6 KU501358 766 Micromonospora peucetia DSM 43363 (X92603.1) 99.83 RKMT_111 (3) D; L DS 6 KU501349 824 Verrucosispora maris AB-18-032 (NR_074617.1) 100 RKMT_117 (2) L H 1 KU501345 812 Cellulosimicrobium funkei W6122 (AY501364.1) 100 RKMT_150 (3) D; L H 5 KU501347 818 Micromonospora tulbaghiae TVU1 (EU196562.2) 100 RKMT_152 (8) D; L H 5 KU501348 823 Micromonospora aurantiaca ATCC 27029 (NR_074415.1) 100 RKMT_158 (3) L H 6 KU501346 825 Rhodococcus pyridinivorans PDB9 (AF173005.1) 100 RKMT_160 (1) D DS 5 KU501359 785 Micromonospora yangpuensis FXJ6.011 (GU002071.2) 100 RKMT_188 (1) O H 1 KU501360 827 Micrococcus luteus NCTC 2665 (NR_075062.2) 100 Firmicutes (34d) RKMT_002 (1) O U MA KU501397 841 Exiguobacterium aestuarii_TF-16 (AY594264.1) 100 RKMT_020 (2) O U MA KU501366 842 Planococcus plakortidis AS/ASP6 (II) (JF775504.1) 99.81 RKMT_032 (1) L U MA KU501381 829 Bacillus pakistanensis NCCP-168 (AB618147.1) 99.05 RKMT_038 (1) L U MA KU501398 606 Jeotgalicoccus aerolatus MPA-33 (GU295939.1) 100 RKMT_043 (2) L U MA KU501367 859 Bacillus aerius 24K (JX009139.1) 100 RKMT_045 (1) L U MA KU501382 785 Salinicoccus jeotgali S2R53-5 (DQ471329.1) 100 RKMT_052 (1) L U MA KU501383 843 Bacillus halmapalus DSM 8723 (X76447.1) 99.81 RKMT_061 (7) D; L; O DS 1 KU501377 845 Bacillus licheniformis ATCC 14580_(KJ826589.1) 100 RKMT_066 (7) D; O DS 1 KU501369 832 Bacillus cibi_JG-30 (AY550276.1) 100 RKMT_078 (1) D DS 2 KU501384 789 Paenibacillus pinisoli NB5 (KC415175.1) 99.43 RKMT_086 (1) D DS 2 KU501385 861 Bacillus flexus NBRC 15715 (AB680944.1) 100 RKMT_087 (3) D DS 2 KU501372 862 Bacillus licheniformis ATCC 14580 (KJ826589.1) 100 RKMT_095 (1) D DS 2 KU501386 845 Bacillus nanhaiisediminis NH3 (GQ292773.1) 99.62 62

RKMT_106 (3) D DS 5 KU501379 832 Bacillus hunanensis JSM 081003 (HM054473.1 100 RKMT_112 (1) L H 1 KU501387 859 Bacillus algicola AB423f (FR775437.1) 100 RKMT_113 (2) L H 1 KU501373 843 Bacillus subterraneus DSM13966T (FR733689.1) 100 RKMT_118 (1) L H 1 KU501388 787 Oceanobacillus iheyensis HTE831 (AB010863.2) 100 RKMT_142 (1) L H 2 KU501389 816 Bacillus vietnamensis 15-1 (AB099708.1) 100 RKMT_148 (2) L H 3 KU501374 586 Lactococcus garvieae JCM 10343 (AB598994.1) 100 RKMT_153 (1) L H 5 KU501390 828 Bacillus hunanensis JSM 081003 (HM054473.1) 99.81 RKMT_157 (2) L H 6 KU501380 847 Bacillus lehensis MLB2 (AY793550.3) 99.62 RKMT_161 (1) O H 3 KU501391 653 Enterococcus gallinarum NBRC 100675 (AB681218.1) 100 RKMT_162 (1) O H 2 KU501392 844 Bacillus litoralis SW-211 (AY608605.1) 100 RKMT_164 (5) D; O H 2 KU501370 844 Fictibacillus barbaricus V2-BIII-A2 (AJ422145.1) 100 RKMT_165 (2) O H 2 KU501375 863 Macrococcus caseolyticus ATCC 13548 (D83359.1) 100 RKMT_172 (1) O H 6 KU501393 846 Bacillus plakortidis P203T (AJ880003.1) 100 RKMT_174 (2) O H 2 KU501378 861 Bacillus barbaricus V2-BIII-A2 (AJ422145.1) 100 RKMT_178 (1) O P 2 KU501394 843 Bacillus idriensis SMC 4352-2 (AY904033.1) 100 RKMT_183 (1) L P 2 KU501395 788 Gracilibacillus ureilyticus MF38 (EU709020.1) 99.81 RKMT_184 (3) L P 2 KU501368 789 Halobacillus litoralis SL-4 (X94558.1) 100 RKMT_185 (3) L P 2 KU501376 790 Halobacillus kuroshimensis IS-Hb7 (AB195680.1) 99.81 RKMT_192 (2) O H 1 KU501365 833 Bacillus oceanisediminis H2 (JX009140.1) 100 RKMT_193 (3) D; L; O H 1 KU501371 844 Bacillus vietnamensis 15-1 (AB099708.1) 100 RKMT_199 (1) L P 1 KU501396 790 Psychrobacillus psychrodurans 68E3 (AJ277984.1) 99.05 Proteobacteria (8d) RKMT_007 (1) O U MA KU501402 851 Psychrobacter piscatorii T-3-2 (AB453700.1) 99.82 RKMT_011 (4) O U MA KU501399 845 Psychrobacter maritimus Pi2-20 (AJ609272.1) 100 RKMT_028a (1) L U MA KU501403 843 Pseudomonas chloritidismutans AW-1 16S (AY017341.1) 99.63 RKMT_050 (1) L U MA KU501404 647 Psychrobacter fulvigenes KC 40 (AB438958.1) 99.08 RKMT_067 (1) D DS 1 KU501406 789 Wenxinia marina HY34 (DQ640643.1) 99.45 RKMT_070 (2) D DS 1 KU501401 731 Erythrobacter vulgaris 022 2-10 (AY706935.1) 100 RKMT_126 (1) W DDC2 MA KU501405 782 Vibrio kanaloae LMG 20539 (AJ316193.1) 100 RKMT_127 (9) O; W DDC2 MA KU501400 781 Pseudoalteromonas undina NBRC 103039 (AB681919.1) 100 aNumber of isolates represented by the representative strains bSource: O - Snail; L - Sea Star; D - Dry Sediment; W - Wet Sediment cPretreatment: U - Untreated; DS - Dry-stamp; H - Heat; P - Phenol; DDC2 - Dispersal and Differential Centrifugation dNumber of OTUs

4.2.1.3. Antimicrobial assay

To screen isolates for their ability to produce antimicrobial substances, all isolates (n=134) were cultivated in marine broth and the fermentation extracts tested for antimicrobial activity against six pathogens. No activity was observed against the Gram-negative pathogens (P. aeruginosa, P. vulgaris) or C. albicans. Extracts from four strains inhibited the growth of the Staphylococcus sp. Two Actinobacteria (Streptomyces sp. and Micromonospora sp.) exhibited strong activity (> 80% growth inhibition) against S. warneri and MRSA. The other two isolates exhibiting antimicrobial activity were Firmicutes (Bacillus sp. and Halobacillus sp.) and showed activity between 60% and 80% growth inhibition (Table 5).

Table 5. Antimicrobial activity of fermentation extracts derived from four strains. No activity was observed against the other four pathogens (E. faecium, P. aeruginosa, P. vulgaris and C. albicans). No antimicrobial activity was observed in extracts from other strains examined Pathogen Marine Isolate S. aureus ATCC 33591 (MRSA) S. warneri ATCC 17917 RKMT_178 (Bacillus sp.) + - RKMT_160 (Micromonospora sp.) ++ - RKMT_184 (Halobacillus sp.) - + RKMT_071 (Streptomyces sp.) - ++ Labels: - No Activity, + 60%<80% activity, ++ >80% inhibition growth

4.2.1.4. Chemical analysis of antibacterial extracts

To identify metabolites potentially responsible for the observed antimicrobial activity, extracts were analyzed by UPLC-PDA-ELSD-HRMS. Peaks with high intensity in the ELSD and/or HRMS chromatograms were selected for further examination as these abundant compounds would be the most likely components of the extract responsible for the observed bioactivity. The mass spectra were analyzed to identify the pseudomolecular ions corresponding to each peak. In all cases two or more pseudomolecular ions ([M+H]+, + + [M+NH4] , [M+Na] ) were observed. To putatively identify these compounds the molecular weights of observed pseudomolecular ions were used to search bacterial metabolites contained in the Antibase database (Wiley®, 2014). Streptomyces sp. RKMT_071 produced four major compounds which eluted at 3.07 min, 4.70 min, 6.83 min and 7.52 min (Figure S1 – Page 101). The peak at 5.40 min in the ELSD trace corresponds to compounds that do not ionize in the positive mode in the mass spectrometry analysis. This fact and the high retention 64

time suggest it corresponds to fatty acids already reported for bacteria (O’LEARY, 1962). The compounds eluting at 3.07 min (m/z 560.3184 [M+H]+) and 4.43 min (m/z 1147.6409 [M+Na]+) had no hits within 5 ppm in the Antibase 2014 database, suggesting these compounds may be novel metabolites. The compound eluting at 4.70 min with a m/z of 479.2908 [M+H]+ had four hits in Antibase with theoretical [M+H]+ masses < 5 ppm from the observed mass (actual deviation 0.75 ppm). All compounds had the molecular formula

C29H38N2O4 and corresponded to the known Streptomyces metabolites ikarugamycin (JOMON et al., 1972), and three acylated phenazines (saphenic acid myristoyl ester, tetradecanoic acid saphenyl ester, and 12-methyltridecanoic acid saphenyl ester) (LAURSEN; NIELSEN, 2004). The same ion was detected in the PDA detector and presented two maximal peaks at 227 nm and 327 nm (Figure S2 – Page 101). Phenazines show a UV absorption at 250 nm and 360 nm (MEHNAZ et al., 2009) and ikarugamycin a UV absorption of 220 nm and 325 nm (JOMON et al., 1972). Therefore, the compound detected is most likely to be ikarugamycin. The compound eluting at 6.83 min corresponded to a compound with a m/z 1111.6389 [M+H]+. Due to the large size of this molecule the search window was broadened to 10 ppm and resulted in a single hit corresponding to valinomycin (Δppm = 5.48) (LI et al., 2015). The extract from Micromonospora sp. RKMT_160 contained two major metabolites (Figure S3 – Page 102). The compound eluting at 1.77 min with a m/z 261.1234 [M+H]+ matched four previously reported compounds in Antibase with [M+H]+ molecular forumulae corresponding to C14H16N2O3 (Δppm = 0.10). These compounds were cyclo(Tyr-Pro), previously reported as produced by Aspergillus flavipes and Streptomyces sp. strains (BARROW; SUN, 1994; WATTANA-AMORN et al., 2015), and cyclo(Phe-4-hydroxyPro), reported as produced by Aureobasidium pullulans and unclassified marine bacteria strains (FDHILA et al., 2003; SHIGEMORI et al., 1998), both diketopiperazines with varying stereochemical configurations. The compound eluting at 2.19 min with a m/z 284.1394 [M+H]+ matched a single entry in Antibase, tryptophan-dehydrobutyrine diketopiperazine

(Δppm = 0.14) (KAKINUMA; KENNETH; RINEHART, 1974). The diketopiperazines with masses matching those of the compounds in the RKMT_160 extracts were all previously reported from Streptomyces. Study of the extract from strain RKMT_178 showed no known compounds produced by this isolate with two major compounds with retention times at 2.52 min and 2.54 min (Figure S4 – Page 102). Halobacillus sp. RKMT_184 produced six major compounds which eluted at 1.97 min, 2.07 min (two compounds), 2.13 min, 2.26 min and 2.28 min. The peaks at 5.10 min and 65

5.57 min in the ELSD trace likely correspond to fatty acids as these compounds did not ionize in positive mode in the HRMS analysis and eluted late in the chromatogram indicating the compounds were highly non-polar (O’LEARY, 1962) (Figure S5 – Page 103). The compounds eluting at 2.28 min (m/z 349.1791 [M+H]+), 2.26 min (m/z 375.1755 [M+H]+) and 2.07 min (m/z 325.1252 [M+H]+) had no hits within 5 ppm in the Antibase 2014 database, suggesting these compounds may be novel metabolites. The compound eluting at 1.97 min with a m/z of 316.1292 [M+H]+ had two hits in Antibase with the molecular formula

C16H17N3O4 (Δppm = 0.09 ppm). The compounds corresponded to the known Streptomyces metabolites anthramycin and maremycin. Due to the antimicrobial activity of the RKMT_184 extract, the detected compound is most likely anthramycin, as this compound has reported antibiotic activity (KOHN; BONO; KANN, 1967). The compound eluting at 2.07 min with a + m/z 351.1213 [M+H] had six hits in Antibase with the molecular formula C21H18O5 (Δppm = 4.04 ppm). However, only one of them, brasiliquinone C, was reported from a bacterial source. Brasilquinone C is active against Gram-positive bacteria, thus may be in part responsible for the antimicrobial activity observed in the RKMT_184 extract (NEMOTO et al., 1997). The compound eluting at 2.13 min with a m/z of 217.0972 [M+H]+ had four hits in

Antibase with the molecular formula C12H12N2O2 (Δppm = 0.54 ppm) which is consistent with the known actinomycetes metabolites N-acetyl-β-oxotryptamine, mansouramycin A, sannanine and caerulomycin F. Since all were previously reported as produced by bacterial sources no specific compound can be attributed to the ion detected. All the extracted chromatograms are shown in Figure S6 – Page 103. A summary of the ions detected in bioactive extract is provided in Table 6.

Mass t Molecular Possible Strain ID R PM ion m/z Error (min) Formula Compound (ppm) [M+H]+ 560.3185 3.07 No hits -- Putatively new [M+Na]+ 582.3002 [M+H]+ 479.2908 4.70 C H N O 0.75 Ikarugamycin [M+Na]+ 501.2729 29 38 2 4 RKMT_071 [M+H]+ 1111.6445 Streptomyces sp. + 6.83 [M+NH4] 1128.6685 C54H90N6O18 5.48 Valinomycin [M+Na]+ 1133.6243 [M+NH ]+ 1142.6852 4.43 4 No hits -- Putatively new [M+Na]+ 1147.6410 [M+H]+ 261.1234 cis- 1.77 C H N O 0.10 RKMT_160 [M+Na]+ 283.1054 14 16 2 3 Cyclo(tyrosylprolyl) Micromonospora [M+H]+ 284.1394 Tryptophan- sp. 2.19 C16H17N3O2 0.14 dehydrobutyrine [M+Na]+ 306.1213 diketopiperazine [M+H]+ 345.1842 2.52 No hits -- Putatively new RKMT_178 [M+Na]+ 363.1948 Bacillus sp. [M+H]+ 389.1911 2.54 No hits -- Putatively new [M+Na]+ 411.1732 [M+H]+ 316.1292 1.97 C H N O 0.09 Anthramycin [M+Na]+ 338.1109 16 17 3 4 [M+H]+ 351.1213 2.07 C H O 4.04 Brasiliquinone-C [M+Na]+ 373.1032 21 18 5 [M+H]+ 325.1252 2.07 No hits -- Putatively new RKMT_184 [M+Na]+ 347.1070 Halobacillus sp. [M+H]+ 217.0973 Four possible 2.13 C H N O 0.54 [M+Na]+ 239.0802 12 12 2 2 compounds [M+H]+ 349.1791 2.28 No Hits -- Putatively new [M+Na]+ 371.1613 [M+H]+ 375.1755 2.26 No hits -- Putatively new [M+Na]+ 397.1574

4.2.2. Metabolomic study of the marine actinomycete Verrucosispora sp.

4.2.2.1. Identification of bacterial strains

Pairwise distance estimation analysis showed 99.78% and 99.93% of similarity when comparing RKMT_073 to RKMT_111 and RKMT_176 sequences (1346 bp), respectively. Considering a cut-off value of 99%, it is possible to confirm that all isolates belong to the same species. Comparison with reference strains of all Verrucosispora species showed a 99.10% of similarity between RKMT_073 and Verrucosispora maris (type strain, GenBank 67

accession nº AY528866.1), the highest value among all reference strains. Thus, pairwise distance analysis showed that all isolates belong to Verrucosispora maris species.

4.2.2.2. Bacteria fermentation and metabolite production evaluation

After a very slow growth in room temperature, evaluation of the strains growth development at 30 ºC revealed a much better growth. Thus, all fermentations were carried out in incubators shaking at 200 rpm and 30 ºC for 7 days.

4.2.2.3. Metabolite production of the isolated strains

To investigate the metabolite production of the isolated strains in different cultivation media, all three strains were cultivated in nine different media and their ability of compounds production monitored by LC-HRMS, cluster and PCA analysis. All extracted compounds were detected within the m/z range of 200-2000 in positive mode. In the PCA plots (Figure 8) it was analyzed any outliers for recognition of specific conditions where production of unique compounds appear as well which are the best conditions for the production of known bioactive metabolites. Therefore it is important to highlight that the PCA was used as a discovery tool and not a statistical way to explain a system. Score plots provided information about which strains in which media stand out as compound producers whether loading plots indicate which compounds are unique in the given conditions. In all PCA analysis replicates fell overlaid confirming the consistency of the data generated.

1 A MB

ISP3m 2

Other media

B Other media Other ions 11 12

BFM-11m BFM-3m 1 13 BFM-5m BFM-4m 10

5 7 2 9 BFM-1m 6 8 3 4

C ASW-Am 1 BFM-11m 10 6 BFM-1m 8 4 9 7 5 3

Other Media 2 Other ions

D RKMT_111 – BFM-1m RKMT_176 – BFM-1m

RKMT_176 – BFM-11m

RKMT_111 – ASW-Am RKMT_111 – BFM-11m

RKMT_176 – ASW-Am

4.2.2.4. Compilation of the data

All the ions detected and their retention times are compiled in Table S1 in Supplementary Material. The table comprises the adducts attributed to the ions, the molecular formula of the compounds, the theorical molecular mass for the corresponding molecular formula and the Δppm error, the corresponding number in the loadings plot for the individual and grouped PCA analyses, the number of Antibase 2012 Wiley® hits (total, bacterial and 69

Verrucosispora sources), the identification of compounds when possible and the media where the compounds were produced. In dark pink it is highlighted the possible analogs and their differences.

4.2.2.5. RKMT_073 metabolites production

A two component PCA model was generated and explains 100% of the variance within the dataset. The small number of components is due to the low number of ions detected for this strain, which does not mean it not produces compounds but it may not be producing in enough amount to be detected in the given analysis parameters. Figure 8-A shows the scores and loadings plots for the first two components of the model. Analysis of the scores plot revealed that cultivation of RKMT_073 in media ISP3m and MB led to the production of unique compounds, since they appear far from 0 and well separated from the majority of samples. Other samples are all overlaid, meaning that they present the same ions or, in this case, no compound production evidenced by no corresponding points in the loadings plot. The compounds detected are shown in Table S1 – Page 108, where each compound is discriminated in which point of the loadings plot is represented. Loadings plot reveals the key buckets responsible for the groupings observed in the scores plot, and consequently secondary metabolites produced in specific media. Since no secondary metabolites were detected in the majority of the media, nothing is observed in the loadings plot corresponding to “other media”. In MB, three unknown compounds were produced consistently in both replicates and are represented by 1.The most prolific media was ISP3m, where six unknown compounds were observed and are represented by 2. None of the ions observed correspond to any of the known compounds previously reported for Verrucosispora sp., but the ion [M+H]+ = 533.3816 detected, represented by 2 in the loadings + plot, was identified as Virustomycin FD-892, [M+H] = 533.3842, Δppm = 4.97 ppm (Antibase), previously reported as produced by another actinobacteria genus, Streptomyces (Table S1 – Page 108 in Supplementary Material).

4.2.2.6. RKMT_111 metabolites production

A four component PCA model was generated and explains 76% of the variance within the dataset. Figure 1-B shows the scores and loadings plots for the first two components of the model. Replicates fell overlaid confirming the consistency of the data generated. Analysis of 70

the scores plot revealed that cultivation of RKMT_111 in media BFM-1m and BFM-11m provided the greater number and diversity of compounds, since they appear further from 0. BFM-3m, BFM-4m and BFM-5m also show good production of unique compounds but in a smaller diversity, since they are closer to 0. Other media showed similar chemical profile and are labeled as “other media” in the loadings plot. Strain RKMT_111 shows a great capability of secondary metabolite production. This is observed by the fact that from nine media utilized, six showed unique ions for this strain. In the loadings plot, 1 corresponds to eight compounds that are exclusively produced in BFM- 3m and BFM-5m. Compounds represented by 2, 3, 4 and 5 are not specific of any media since all of them are produced in BFM-1m, wherein 2 corresponds to four compounds that are also produced in BFM-3m and BFM-5m, 3 to one compound also produced in BFM-4m and BFM-5m, 4 to four compounds also produced in BFM-4m and 5 to one compound also produced in BFM-2m. In 6 there are represented twenty compounds, all produced in BFM-1m exclusively, where it was detected the ion [M+H]+ = 377.19583 which may be attributed to + the substance 16α-Hydroxyprednisolone, [M+H] = 377.1959, Δppm = 0.10 ppm, already reported as produced by bacteria (Scifinder). Compounds represented in 7, 8, 9 and 10 are all produced in BFM-1m and BFM-11m, wherein one compound is represented by 7 and is also produced in BFM-4m and MB, one compound is represented by 8 and is also produced in BFM-4m and five compounds are represented by 9, where it was detected the ions [M+H]+ = 257.1172 which may be attributed + + to Kurasoin A, [M+H] = 257.1172, Δppm = 0.19 ppm (Scifinder) and [M+H] = 231.1016, attributed to Methyl-3-methoxy-5-methyl-naphthalene-1-carboxylate, [M+H]+ = 231.1016,

Δppm = 0.17 ppm (Antibase), both previously reported as produced by bacteria. One compound is represented by 10 and is also produced in BFM-2m. Compounds represented in 11, 12 and 13 are all produced in BFM-11m, wherein one compound is represented by 11 and is also produced in BFM-2m, three compounds are represented in 12 and is also produced in ISP3m and forty-eight compounds are represented by 13, all produced exclusively in BFM-11m. In 13, it was detected the ion [M+H]+ = 247.1082 which + may be attributed to Anthramycin, [M+H] = 247.1077, Δppm = 2.02 ppm (AntiBase). It was also detected the adducts [M+H]+ = 223.0964 attributed to Talomone, [M+H]+ = 223.0964, + Δppm = 0.36 ppm; [M+H] = 219.1017 correspond to 4H-1,3-Benzodioxin-4-one,2,2-dimethyl- + + 5-(2-propen-1-yl), [M+H] = 219.1016, Δppm = 0.38 ppm (Scifinder); [M+H] = 293.1383 attributed to 4H-1-Benzopyran-4-one,2-butyl-8-hydroxy-5,7-dimethoxy-3-methyl, [M+H]+ = + 293.1384, Δppm = 0.27 ppm (Scifinder); [M+Na] = 591.2205 attributed to Chinikomycin A, 71

+ + [M+Na] = 591.2232, Δppm = 4.70 ppm (Antibase); [M+Na] = 505.1833 attributed to + Hydroxystaurosporine, [M+Na] = 505.1846, Δppm = 2.58 ppm (Antibase). All compounds attributed to the ion adducts detected were previously reported as produced by bacteria. The region labeled as “Other ions” do not contribute much in the PCA, meaning that the ions represented by those components are produced in the majority of the cultivation media tested. In this case, 82 ions are represented in this region. Among them it was detected + + the ions [M+H] = 393.2727 attributed to Fortimicin-AS, [M+H] = 393.2708, Δppm = 4.89 + + ppm and [M+H] = 521.3475 attributed to Butyrolactol, [M+H] = 521.3473, Δppm = 0.47 ppm, both previously reported as produced by bacteria (Antibase). Among all secondary metabolites previously reported for Verrucosispora sp., only in BFM-11m it was possible to identify known compounds. In this media, the ion [M+H]+ = 347.1490 was detected, corresponding to the known Abyssomicin C, [M+H]+ = 347.1489, + Δppm=0.13 ppm. The ion [M+H] = 349.1646 was also detected and correspond to either of + the isomers Abyssomicin D or Abyssomicin H, [M+H] = 349.1645, Δppm = 0.07 ppm. The abyssomicins detected are represented in 11 and 13 in the loadings plot.

4.2.2.7. RKMT_176 metabolites production

A three component PCA model was generated and explains 66% of the variance within the dataset. Figure 1-C shows the scores and loadings plots for the first two components of the model. Analysis of the scores plot revealed that cultivation of RKMT_176 in media BFM-1m, BFM-11m and ASW-Am led to the production of unique compounds for these strains in these media, since they appear far from 0 and well separated from the majority of samples. Strain RKMT_176 showed a great ability to produce secondary metabolites in ASW- Am media, where up to 40 unique compounds were detected and are represented by 1 in the loadings plot. Compounds represented by 2, 3, 4, 7, 8, 9 and 10 are all produced in BFM-1m, wherein one compound is represented by 2, which is also produced in BFM-3m and BFM- 5m, one compound is represented by 3 and also produced in BFM-4m and eleven compounds are represented by 4 where it was detected the adduct ion [M+H]+ = 275.1277 attributed to + Diolmycin, [M+H] = 275.1278, Δppm = 0.37 ppm (Antibase). One compound is represented by 7 and is also produced in BFM-3m and BFM-11m, one compound is produced by 8 and is also produced in BFM-11m and MB and two compounds are represented by 9 and are also produced in BFM-3m and BFM-11m, where it was detected the compound Methyl-3- 72

methoxy-5-methyl-naphthalene-1-carboxylate also produced by RKMT_111 (Figure S7 – Page 104). Seventeen compounds are represented by 10 and are also produced in BFM-11m where it was detected the adduct ion [M+H]+ = 335.1490 attributed to (-)-Tetrodecamycin, + [M+H] = 335.1489, Δppm = 0.18 ppm (Antibase). In 10 it was also identified the compound Kurasoin A also produced by RKMT_111 (Figure S8 – Page 104). Compounds represented by 5 and 6 are produced in BFM-11m, wherein one compound is represented by 5 and is also produced in MB and seventeen compounds are represented by 6, where it was detected the compounds Anthramycin, Talomone, 4H-1,3-Benzodioxin-4-one,2,2-dimethyl-5-(2-propen-1- yl) and 4H-1-Benzopyran-4-one,2-butyl-8-hydroxy-5,7-dimethoxy-3-methyl also produced by RKMT_111 (Figures S9-S12, respectively – Pages 105-106). The region labeled as “Other ions” do not contribute much in the PCA, meaning that the ions represented by those components are produced in the majority of the cultivation media tested. In this case, 58 ions are represented in this region where it was detected the adduct ion [M+H]+ = 479.2864 + attributed to Terragine D, [M+H] = 479.2864, Δppm = 0.10 ppm (Antibase). The compound Butyrolactol is represented in “Other ions” and was also produced by RKMT_111 (Figure S13 – Page 107). All compounds identified were previously reported as produced by bacterial isolates.

4.2.2.8. Comparison of metabolites production between strains

A seven component PCA model was generated and explains 67% of the variance within the dataset. Figure 1-D shows the scores and loadings plots for the first two components of the model. Analysis of the scores plot revealed that cultivation of RKMT_111 and RKMT_176 in media BFM-1m, BFM-11m and ASW-Am led to the production of unique compounds for these strains in these media, since they appear far from 0 and well separated from the majority of samples. The scores plot shows that RKMT_111 and RKMT_176 possess the highest capability of compound production in BFM-1m, BFM-11m and ASW-Am. It is also possible to see that both strains present similar ions when cultivated in the same media but also show unique ions. This fact is acknowledge in the analysis of the loadings plot, where 20 compounds that are produced by both strains in ASW-Am are represented by 1, while 23 compounds that are produced exclusively by RKMT_176 in the same medium are represented by 2 and 3 compounds that are produced exclusively by RKMT_111 are represented by 3, also in ASW- 73

Am. The same is observed for the cultivation of those strains in BFM-1m and BFM11m. Both strains RKMT_111 and RKMT_176 produced 18 compounds that are represented in 4, 5, 6, 7, 8, 9, 18, 19, 21, 22 and 23. Those compounds do not fall overlaid because they may also be produced in different other media. In 10, 11, 12 and 13 there are represented 18 compounds produced in BFM-11m by both strains, with some compounds being produced in other media by one of the strains. There are 62 compounds produced exclusively by RKMT_111 which are represented by 15, 16, 24 and 25, where forty-five compounds are represented by 15 and are produced exclusively in BFM-11m and fifteen compounds are represented by 24 and are exclusively produced in BFM-1m. The compounds produced exclusively by RKMT_176 are represented by 14, where nine compounds are represented and produced in BFM-1m and BFM-11m. In 17, nine compounds are represented and produced in BFM-11m and in 20, 7 compounds are represented and produced in BFM-1m. It is easily noticed that media BFM-1m and BFM-11m contain the best compositions for the production of unique compounds by the strains tested. Also, even that both strains show similar compounds, there were observed unique ions for both strains. The strain RKMT_073 did not show any extra difference compared to the other two, meaning that the compounds produced by this strain were also found in the others. The region labeled as “Other ions” do not contribute much in the PCA, meaning that the ions represented by those components are produced in the majority of the cultivation media tested. In this case, 107 ions are represented in this region.

4.2.2.9. Cluster analysis and barcode

Cluster analysis of all three strains combined with the binary barcode generated the scheme in Figure 9.

111111 1111111 1 1 111 1 1111 1111 1111 1 1 1 1 1 1 111111 1111111 1 1 111 1 1111 1111 1111 1 1 1 1 1 1 1111 11111111 1 11111 1 1 1 1 11111111111111 11 1 1 1 1 1 1 1 1 1 1111 11111111 1 11111 1 1 1 1 11111111111111 11 1 1 1 1 1 1 1 1 1 1111111 11111 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 11 11111111 1111111 11111 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 11 11111111 1111 11111111 1 1 1 1 1 11111111111 1 111111111111111111111111111111111111111111111 1111 11111111 1 1 1 1 1 11111111111 1 111111111111111111111111111111111111111111111 1 1 1 1 1 111111111111111 1 11111 11111 11 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 111111111111111 1 11111 11111 11 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 111111111111111 1 1 1 1 1 1 1 1 111111111111111 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Figure 9. Chemical barcoding and cluster analysis of the three strains of Verrucosispora maris in all media tested

The barcode generated was organized for representation of the chemical diversity from the smaller retention time to the highest (left to right). Compounds represented in the barcode were consistently detected in both replicates. Cluster analysis dendogram shows the similarity in the production of compounds of the tree strains in the nine media tested according to their chemical fingerprint. Also, cluster analysis indicates the chemical diversity provided by the bacterial isolates. Strains RKMT_111 and RKMT_176 in media BFM-1m and BFM-11m appear in the top of the dendogram with the highest number of compounds produced, while RKMT_073 in media BFM-1m, BFM-2m, BFM-3m, BFM-4m, BFM-5m, BFM-11m and ASW-Am appear in the bottom of the dendogram with the lowest number of compounds. No compounds were detected for RKMT_073 in these media, which may not mean there is no secondary metabolites production by this strain but instead the intensity of the ions in the LC- MS analysis may be lower than the threshold chosen for metabolomics. A lower threshold was chosen for analysis but an increase of the noise level was too expressive and an approximate result was obtained regarding the similarity among the samples (data not shown). Cluster analysis clearly shows two main groups. The first one, at the top, comprises the strains RKMT_111 and RKMT_176 in BFM-1m and BFM-11m, where the majority of compounds were observed. Both strains produce several same compounds like the identified 75

anthramycin, talomone, 4H-1,3-Benzodioxin-4-one, 2,2-dimethyl-5-(2-propen-1-yl), 4H-1- Benzopyran-4-one,2-butyl-8-hydroxy-5,7-dimethoxy-3-methyl and Methyl 3-methoxy-5- methyl-naphthalene-1-carboxylate in BFM-11m, kurasoin A in BFM-1m and BFM-11m and butyrolactol in BFM-4m (see Chromatograms in Supplementary Material). Also in this group are the abyssomicin C and abyssomicin D or H, produced in BFM-11m and chemical markers of the genus. The second group comprises the majority of the samples with some not identified compounds being produced by both strains RKMT_111 and RKMT_176 but mainly compounds specific of each strain/medium.

4.2.3. Biotransformations in bacterial isolates

4.2.3.1. Identification of the bacterial strain

Phylogenetic relationship of RKMT_070 with reference strains that showed highest similarity after BLAST search obtained from Genbank is shown in Figure 10.

Pairwise distance estimation analysis showed 99.99% of similarity between RKMT_070 and type strain Erythrobacter vulgaris 022 2-10 (AY706935.1) confirming this is 76

the most likely species of the isolate. BLAST search in Genbank also showed the same strain as presenting the highest identity level compared with RKMT_070. Phylogenetic analysis did not grouped exactly both sequences but this can be explained by the fact that the bootstrap values were <50%, meaning a high similarity between RKMT_070, Erythrobacter vulgaris 022 2-10 (AY706935.1) and Erythrobacter flavus SW-46 (AF500004.1).

4.2.3.2. Comparison of samples by LC-HRMS analysis

Analysis of samples showed high similarity between corresponding samples from cultivation in tubes and baffled flasks (Figure S14 – Page 117). Fractions from HP-20 extracted with MeOH showed the higher number of substances in greater amount, indicated by the ELSD detector (Figure S14 – Page 117). Therefore both MeOH fractions from HP-20 were combined and fractionated by HPLC-MS (Figure S15 – Page 117).

4.2.3.3. Isolation and identification of compounds

Fractionation by HPLC-MS yielded twelve fractions which were analyzed by HPLC- PDA-ELSD-HRMS. Fractions 01, 04, 09, 10 and 12 did not show expressive peaks in the ELSD detector and were not considered for further analysis. Fractions 02, 03, 05, 06 and 08 were pure compounds also present in the medium blank analysis and were identified based on the HRMS data obtained (Figure S16 – Page 118). Fraction 02 showed one compound eluting + at 2.90 min (m/z 466.3153 [M+H] ) which corresponds to glycoholic acid (Δppm = 3.43). Fraction 03 showed one compound eluting at 3.01 min (m/z 407.2791 [M+H]+) which corresponds to 7-oxo-cholic acid (Δppm = 1.72). Fraction 05 showed one compound eluting at + 3.23 min (m/z 409.2945 [M+H] ) which corresponds to cholic acid (Δppm = 2.20). Fractions

06 (rt = 3.23 min) and 08 (rt = 3.30 min) showed isomers with exact same m/z 450.3214, which may correspond to glycodeoxycholic acid and glycochenodeoxycholic acid (Δppm = 1.33). Compounds detected were previously described as constituents of the peptone present in the composition of the media utilized (KAMEKURA et al., 1988). Fractions 07 and 11 + were pure compounds (m/z 508.3262 [M+H] , C28H45NO7, Δppm = 2.36, and m/z 492.3322 + [M+H] , C28H45NO6, Δppm = 0.61, respectively) not detected in the medium blank (Figure S17 – Page 118) and with no hits in the Antibase 2012. Chromatographic data of the fractionation and analysis by LC-HRMS of the fractions is compiled in Table 7. 77

Table 7. Chromatographic data of the fractionation and analysis by LC-HRMS of the fractions collected r (min) r (min) Molecular Mass Error Fraction t t PM ion m/z Compound (HPLC-MS collection) (UPLC-HRMS) formula (Δppm) [M+H]+ 1 0.00-8.00 No detected peaks – – – – – – – – [M+Na]+ [M+H]+ 466.3153 3.43 2 8.80 2.90 C26H43NO6 Glycoholic acid [M+Na]+ 488.2969 2.87 [M+H]+ 407.2791 1.72 3 11.40 3.01 C24H38O5 7-oxo-cholic acid [M+Na]+ 429.2611 1.40 [M+H]+ 4 12.00-16.00 No detected peaks – – – – – – – – [M+Na]+ [M+H]+ 409.2945 2.20 5 16.80 3.18 C24H40O5 Cholic acid [M+Na]+ 431.2766 1.62 [M+H]+ 450.3214 1.33 6 19.30 3.23 C26H43NO5 Glycodeoxycholic acid [M+Na]+ 472.3032 1.48 [M+H]+ 508.3262 2.36 7 20.30 3.25 C28H45NO7 Not reported [M+Na]+ 530.3077 3.20 [M+H]+ 450.3214 1.33 8 20.95-21.80 3.30 C26H43NO5 Glycochenodeoxycholic acid [M+Na]+ 472.3031 1.69 [M+H]+ 9 22.25 No detected peaks – – – – – – – – [M+Na]+ [M+H]+ 10 24.10 No detected peaks – – – – – – – – [M+Na]+ [M+H]+ 492.3322 0.61 11 25.26 3.81 C28H45NO6 Not reported [M+Na]+ 514.3139 1.17 [M+H]+ 12 25.50-30.00 No detected peaks – – – – – – – – [M+Na]+ 78

Fractions 07 and 11 were analyzed by NMR spectroscopy for fully identification of compounds.

4.2.3.4. NMR analysis

High-resolution mass spectrometry suggested the compounds from fractions 07 and 11 were derivatives from glycocholic and glycodeoxycholic acid, respectively, with an acyl- group (CH3CO-), since a difference of 42 Da was observed. It was confirmed by the detection of two additional carbons (one methyl group and one carbonyl group) in HMQC and HMBC. Compound from fraction 07 was identified as a derivative of glycocholic acid (Table

8). A deshielding effect was observed in the chemical shift of hydrogen in position 3 (δH =

4.48 ppm) compared to glycocholic acid (δH = 3.49 ppm), suggesting a modification in the neighborhood. Therefore it was suggested that an acylation occurred in the oxygen in position 3, since this would cause the change in the chemical shift of the H-3 observed. The same effect was observed in NMR analysis of compound from fraction 11, a derivative of glycodeoxycholic acid (Table 8). The chemical shift of H-3 (δH = 4.61 ppm) was more deshielded compared to H-3 from glycodeoxycholic acid (δH = 3.63 ppm), suggesting that the acylation also occurred at position 3. Therefore compounds isolated in fractions 07 and 11 were identified as 3-acetyl-glycocholic acid and 3-acetyl-glycodeoxycholic acid, respectively (Figure 11). NMR spectra are shown in Figures S18-S23 (Pages 119 – 121) in Supplementary Material.

Figure 11. Structure of compounds isolated from Erythrobacter vulgaris fermentation 79

Table 8. Chemical shifts of glycocholic, glycodeoxycholic (IJARE et al., 2005), 3-acetyl-glycocholic and 3-acetyl-glycodeoxycholic acids (14.0 T, MeOD) Glycocholic acid Fraction 7 – 3-acetyl-glycocholic acid Glycodeoxycholic acid Fraction 11 – 3-acetyl-glycodeoxycholic acid Position δC, type δH (HMQC) δC, type δH (HMQC) HMBC δC, type δH (HMQC) δC, type δH (HMQC) HMBC

1 37.7, CH2 1.00; 1.80 35.2, CH2 1.00; 1.81 38.2, CH2 0.98; 1.81 35.3, CH2 0.99; 1.77 0.91 2 32.1, CH2 1.39; 1.66 26.7, CH2 1.53; 1.60 32.0, CH2 1.42; 1.69 26.5, CH2 1.53; 1.61 3 74.4, CH 3.49 75.6, CH 4.48 2.38 74.2, CH 3.63 74.9, CH 4.61 4 41.2, CH2 1.70; 2.07 35.6, CH2 1.65; 2.38 38.0, CH2 1.54; 1.81 32.7, CH2 1.48; 1.89 5 43.9, CH 1.44 42.1, CH 1.39 44.9, CH 1.44 42.6, CH 1.41 0.91; 1.53 6 36.7, CH2 1.55; 1.97 34.7, CH2 1.47; 1.93 30.0, CH2 1.34; 1.85 34.7, CH2 1.48; 1.84 7 71.1, CH 3.89 68.2, CH 3.76 29.0, CH2 1.23; 1.45 26.5, CH2 1.11; 1.41 8 42.1, CH 1.59 40.2, CH 1.51 38.9, CH 1.44 36.3, CH 1.40 9 29.2, CH 2.12 26.9, CH 2.23 36.2, CH 1.85 33.5, CH 1.88 3.91 10 37.2, C -- 35.7, C -- 0.89 36.8, C -- 35.0, C -- 0.91 11 30.5, CH2 1.59 28.8, CH2 1.54 31.1, CH2 1.55 29.0, CH2 1.48 12 75.9, CH 4.05 73.0, CH 3.92 0.68 75.9, CH 4.05 73.0, CH 3.91 0.67 13 49.0, C -- 46.5, C -- 0.68 49.1, C -- 47.7, C -- 0.67 14 44.4, CH 1.83 42.2, CH 1.95 0.68 50.6, CH 1.61 48.5, CH 1.57 3.91 15 25.7, CH2 1.12; 1.71 23.5, CH2 1.07; 1.70 26.6, CH2 1.07; 1.67 24.0, CH2 1.06; 1,57 16 30.0, CH2 1.29; 1.96 27.7, CH2 1.26; 1.88 30.4, CH2 1.27; 1.95 27.7, CH2 1.25; 1.88 17 49.2, CH 1.74 47.3, CH 1.82 0.68 49.1, CH 1.77 47.5, CH 1.80 0.67; 0.99; 1.57 18 14.9, CH3 0.71 12.1, CH3 0.68 15.4, CH3 0.72 12.3, CH3 0.67 1.57; 1.80 19 24.9, CH3 0.91 22.4, CH3 0.89 25.8, CH3 0.94 22.7, CH3 0.91 20 38.0, CH 1.44 36.1, CH 1.38 38.4, CH 1.44 36.3, CH 1.4 0.99 21 19.5, CH3 1.00 16.9, CH3 0.99 19.5, CH3 1.02 16.9, CH3 0.99 22 34.4, CH2 1.43; 1.77 32.3, CH2 1.32; 1.77 34.4, CH2 1.46; 1.74 32.3, CH2 1.32; 1.76 0.99; 2.12; 2.28 23 35.3, CH2 2.23; 2.36 33.3, CH2 2.11; 2.29 35.3, CH2 2.21; 2.39 33.3, CH2 2.12; 2.28 24 180.0, C -- 175.8, C -- 2.11; 2.28 179.7, C -- 175.9, C -- 2.12; 2.28 25 46.1, CH2 3.74 43.8, CH2 3.69 46.2, CH2 3.75 43.3, CH2 3.70 26 179.5, C -- 175.6, C -- 3.69 179.4, C -- 175.2, C -- 3.70 27 -- -- 172.0, C -- 1.93 -- -- 171.5, C -- 1.94 28 -- -- 28.5, CH3 1.93 -- -- 20.6, CH3 1.94 80

Since the identified compounds were not present in the medium blank and are derivatives of compounds from the composition of the medium utilized for cultivation, it was suggested that a biotransformation occurred and an acylation of glycocholic acid and glycodeoxycholic acid is proposed.

5. Discussion

In Brazilian waters, there are many species of economic interest. Much of this interest relates to fishing for fish and shrimp. However, frequently this economic activity is not performed in a sustainable manner, interfering with biodiversity and the natural equilibrium of ecosystems (BARBRAUD et al., 2013; GAMFELDT et al., 2015). One of the shrimp fishery critical aspects is the by-catch fauna, since many non-target species are also captured, and they are promptly discarded, due to the absence of commercial value of these species. Most of the caught species are fishes and crustaceans, but there are also many other invertebrates, from which very little is known about their biology and, still lesser, from their chemical composition. Therefore, it is essential to carry out an inventory of the animals captured, so that we can define actions that may lead to biodiversity conservation of the Brazilian marine resources. In our preliminary studies about the potential of the by-catch to find promising chemical compounds, we performed six trawling in three localities along the coast of the State São Paulo. We identified 71 fish species, 16 crustacean, 7 condrichtes, 4 commercial shrimps and 6 non-commercial shrimps, 7 mollusks, 6 echinoderms, 4 cnidarians and and one sipuncula. Compared to the other phyla, very few equinoderms were captured in our study, which is in accordance to the literature (SEVERINO-RODRIGUES; GUERRA; GRAÇA-LOPES, 2002; (SEVERINO-RODRIGUES; HEBLING; GRAÇA-LOPES, 2007). Besides, like cnidarians, the capture and transport produced damages in the body of these invertebrates. On the other hand, the best results of our chemical screening were obtained with equinoderms. Therefore the equinoderm Luidia senegalensis was selected for chemical investigation.

5.1. Mass spectrometry study of the starfish Luidia senegalensis

Our results demonstrated that noteworthy characteristics of L. senegalensis are the presence of highly polar asterosaponins with 5 and 6 sugar units. Datta et al. (DATTA; TALAPATRA; SWARNAKAR, 2015) described that sulphated steroidal saponins present several biological activities, like hemolytic, antineoplastic, cytotoxic, antitumor, antibacterial, antiviral, antifungal and anti-inflammatory. Besides, they could be involved in several spheres of living like chemical defense, digestion and reproduction (CLARCK; DOWNEY, 1992). As 82

far as we are concerned, no previous reports on the chemical composition of this starfish was reported. Since that sugar identities and connectivity may vary in starfish saponins, it was not possible to fully identify these compounds in this study. On the other hand, we were not able to find in the literature compounds with the combination of aglycone and sugar sequences described in Table 2. Hence, despite asterosaponins are well-known secondary metabolites of starfish, it is possible that this marine invertebrate presents new molecules, which will be investigated in a near future. In the context of this study, the combination of direct injection ESI-IT-MSn and UPLC-ESI-IT-MSn experiments helped us to support the presence of important compounds in invertebrates associated to the by-catch fauna of the shrimp fishery, using a fast and efficient method. This result can contribute to a more rational and sustainable use of the Brazilian marine biodiversity resources.

5.2. Marine bacteria

5.2.1. Cultivable bacterial communities of marine sediment and invertebrates

The emergence of so-called “superbacteria” is a serious health concern as nosocomial infections by Methicillin-Resistant Staphylococcus aureus (MRSA) strains, as one example, are responsible for 19,000 deaths per year in the United States. These pathogens are resistant to β-lactam antibiotics (e.g. penicillins, cephalosporins, carbapenens), which are an important class of antimicrobial agents, comprising more than 65% of the world antibiotic market (POOLE, 2004). Thus, the discovery of new antibiotics efficacious against drug-resistant pathogens is critically important. As the majority of antibiotics are derived from microorganisms, screening new microorganisms for novel antibiotics is of utmost importance for the discovery of more effective drugs. Brazil hosts approximately 20% of the entire world's macro-organism biological diversity (PYLRO et al., 2014). Despite its geographic size, Brazilian marine microbial diversity is still underexplored (BERLINCK et al., 2004). In order to improve knowledge regarding the biotechnological potential of the Brazilian biodiversity, we investigated the cultivable bacteria associated with two marine invertebrates and a sediment sample collected from the Ubatuba region of the Brazilian coast and tested the ability of these bacteria to produce antimicrobial agents. 83

Comparison of these three sources showed that the sea star yielded the highest number of isolates and greatest diversity of genera, including isolates of Micromonospora, Streptomyces and rare genera such as Serinicoccus and Verrucosispora (YANG et al., 2013; YI-LEI et al., 2014). The sediment provided the second highest number of isolates and diversity of genera, including Micromonospora, Streptomyces and Verrucosispora. The lowest number of isolates and genera was obtained from O. urceus. However, rare bacteria as Serinicoccus sp. were isolated from this source. These genera, together with Streptomyces, are actinomycetes with a proven ability to produce bioactive metabolites (SOLANKI; KHANNA; LAL, 2008). These results showed the potential of by-catch invertebrates as a source of a wide diversity of bacterial genera with biotechnological potential. Of the isolation media employed in this study, media 4 and 5 exhibited the greatest selectivity for the isolation of Actinobacteria. This may be due to the low nutrient concentration of these media, thereby reducing the growth of non-actinomycete bacteria and preventing the overgrowth of actinomycetes (JENSEN et al., 2005). The selectivity for actinomycetes can also be attributed to the use of novobiocin in these media, since it suppresses the growth of Gram-positive bacteria (DALISAY et al., 2013), such as Firmicutes, which were abundant on other media. Marine agar (MA) was also a good medium for the isolation of Actinobacteria, although this non-selective medium favored the cultivation of primarily non-filamentous Actinobacteria. The paucity of filamentous bacteria isolated on non-selective media indicates that filamentous actinomycetes are relatively rare in the samples examined in this study, thus necessitating the application of selective isolation media to increase the frequency of isolating these bacteria. The fewest isolates were obtained from Media 3, which was supplemented with rifampicin and streptomycin to select for the cultivation of rare actinomycetes. The low yield of isolates from this media suggests that the combination of antibiotics inhibited the vast majority of bacteria present in the three samples studied (BAKKER-WOUDENBERG et al., 2005; PANKEY; SABATH, 2004). Two different sets of pretreatments were applied to the sediment and invertebrate samples. The dry-stamp method was superior to the DDC pretreatment both in terms of diversity of genera obtained and selectivity for actinomycetes. Drying the sediments combined with the use of selective media selected against the isolation of Gram-negative bacteria, which are abundant in marine habitats, and clearly favored the growth of Gram- positive bacteria (GONTANG; FENICAL; JENSEN, 2007). The DDC method yielded very few isolates overall and no Actinobacteria despite this approach being successfully applied for the isolation of actinomycetes from soil when coupled with selective isolation media 84

(SEMÊDO et al., 2001). This may be explained by the fact that the method is not selective for Actinobacteria and allows for the growth of fast-growing Proteobacteria that overgrew Actinobacteria present in the sample. The low productivity of the DDC method could also be due to the use of deionized water with a marine sediment. The use of deionized water, rather than sea water, may have caused osmotic stress, resulting in reduced viability of marine bacteria present in the sediment. The heat-shock method was clearly the most productive pretreatment applied to the invertebrate samples. This approach has been reported to control the over growth of non-actinomycete colonies on isolation plates (VARGHESE; JAYASRI; SUTHINDHIRAN, 2015). In our study, the heat-shock method reduced the growth of unwanted bacteria and allowed the unfettered growth of more interesting isolates. Pretreatment with 1.5% phenol has been used for the isolation of sporogenous taxa and members of the less-abundant Streptomyces spp. (HAYAKAWA; YOSHIDA; IIMURA, 2004). The low number of isolates obtained using the phenol pretreatment may be due to a low abundance of sporulating taxa in these samples. Untreated samples were a good source for the isolation of bacteria in general, but with no selectivity as Actinobacteria, Proteobacteria and Firmicutes were isolated. The high proportion of Actinobacteria isolated from untreated samples may be attributed to the wide dilution performed, which allowed actinomycetes to avoid over growth. Screening 134 isolates for antimicrobial activity identified four strains that produced antimicrobial substances active against staphylococci. The relatively low hit rate observed among these isolates is likely due to the use of a single fermentation medium, as the production of secondary metabolites is highly dependent on media composition (BODE et al., 2002). Further exploration of the metabolomes of the strains isolated in this study using a wider array of fermentation conditions would undoubtedly uncover additional bioactivity. The observed bioactivity was limited to four isolates: two Firmicutes (Bacillus and Halobacillus) and two Actinobacteria (Micromonospora and Streptomyces strains). Bacillus spp. are well known for their ability to produce antimicrobial lipopeptides such as gramicidin A which is active against S. aureus (LIOU et al., 2015). Interestingly, no known lipopeptides were identified in the fermentation extracts of RKMT_178 and RKMT_184, however, several low molecular weight compounds were detected in UPLC-HRMS analyses of fermentation extracts from these strains. Compounds anthramycin and brasiliquinone-C were identified in the extract from RKMT_184. Both compounds are reported as antimicrobial agents, and may be responsible for the antimicrobial activity exhibited by the fermentation extract of this strain (KOHN; BONO; KANN, 1967; NEMOTO et al., 1997). N-acetyl-β-oxotryptamine, 85

mansouramycin A, sannanine and caerulomycin F were also tentatively identified, although further study is needed to confirm the identity of the active constituents of these extracts. The other compounds did not correspond to known metabolites present in Antibase 2014. Consequently, one or more of the metabolites produced by these strains may represent novel antimicrobial metabolites. Additionally, it should be noted that these strains were isolated form O. urceus and L. senegalensis, underscoring the biomedical potential of microbes obtained from by-catch organisms. Actinobacteria such as Micromonospora spp. are sources of antibiotics such as aminoglycosides, macrolides and ansamycins (WAGMAN; WEINSTEIN, 1980). The antimicrobial activity observed in fermentation extracts of RKMT_160 can be tentatively associated to the presence of the diketopiperazines cis-cyclo(tyrosylprolyl) and tryptophan- dehydrobutyrine diketopiperazine, previously reported for Streptomyces spp. with antimicrobial activity (ELLEUCH et al., 2010; KAKINUMA; KENNETH; RINEHART, 1974). The occurrence of the same secondary metabolites in both Streptomyces spp. and Micromonospora spp. has been reported previously (WAGMAN; WEINSTEIN, 1980). Also, production of diketopiperazines by Micromonospora spp. has also been reported (YANG et al., 2004). Streptomycetes are the most studied Actinobacteria and are known for the production of a wide diversity of active secondary metabolites such as antibiotics, antivirals, antitumoral, and immunosuppressives. The presence of the potent antibiotic ikarugamycin detected, previously reported as produced by Streptomyces spp. (JOMON et al., 1972; LI et al., 2015), is likely responsible for the high activity observed for the isolated Streptomyces sp. RKMT_071.

5.2.2. Metabolomic study of the marine actinomycete Verrucosispora sp.

Identification of bacterial isolates indicated that all the three strains isolated belong to the species Verrucosispora maris, first described by Goodfellow et al., in 2012 (GOODFELLOW et al., 2012). The genome of the type strain AB-18-032 (GenBank accession nº AY528866.1) was sequenced before and it was found that this strain is a producer of abyssomicins, the first known natural product inhibitor of the para-aminobenzoic acid biosynthetic pathway, and proximicins, which presents novel antitumor properties (ROH et al., 2011). Therefore, it is important to verify whether the strains isolated are capable of 86

producing those compounds, important natural products, since it is known that bacteria of the same species may or may not produce the same metabolites (LU et al., 2014). Another question is what are the best conditions for the production of these metabolites and if isolates from the same species and site collection that were under the same environmental conditions are capable of producing the same compounds. The first question was assessed by the cultivation of the isolates in different culture media, at 30ºC. This temperature was chosen because it was observed no growth on the isolation plates at room temperature, observing the appearance of colonies just after incubation at 30ºC. After cultivation of the isolates in the nine media previously described and LC-HRMS analysis, the strains were separately analyzed by PCA and it was clear that just RKMT_111 was capable of producing the known abyssomicin C and the abyssomicins D and/or H (isomers). Moreover, these compounds were observed only in BFM-11m and in no other media tested. This finding is key for the production of abyssomicins in future studies, where no media composition optimization will be necessary. The total synthesis of abyssomicin C has been achieved by Nicolaou & Harrison in 2007, but its structure presents several challenging elements like a 11-membered macrocyclic ring, seven stereogenic centers, a potentially reactive α,β –unsaturated ketone and a novel fused tetronate oxabicyclo[2.2.2]octane core (NICOLAOU; HARRISON, 2007), which makes the total synthesis laborious and time-consuming. Another report of the total synthesis of an analogue of Abyssomicin C, the (−)-atrop–abyssomicin C, presented a 21 steps synthesis with 1.8% of yield (BIHELOVIC et al., 2013). Thus, production optimization by bacterial strains appears to be a much better option for the obtainment of these compounds. The PCA analysis of metabolite production by the isolates separately revealed that strain RKMT_073 is the less capable of producing secondary metabolites in the given conditions. It only produced compounds in MB (3 compounds) and ISP3m (5 compounds). On the other hand, strain RKMT_111 showed a great ability of compound production, with 153 unique compounds detected in five different media previously described, standing out BFM-1m and BFM-11m. In these media, it was observed 29 and 48 compounds, respectively, a great number comparing to other media. The same media were good for metabolite production for RKMT_176 as it is seen in PCA and cluster analysis. While other media grouped together close to zero, fermentation in BFM-1m led to the production of 23 compounds and in BFM-11m to the production of 30 compounds. Medium ASW-Am was especially good for compound production for strain RKMT_176, where 39 compounds were detected. In brief, BFM-1m and BFM-11m were the best media for production of compounds 87

by the strains tested, with BFM-11m being the only media where the bioactive compounds abyssomicins were detected. When assessing secondary metabolite production of isolates from the same bacterial species, it is always challenging to select strains for cultivation. Even small-scale processes used as tools for strains prioritization as LC-MS based metabolomics (HOU et al., 2012) and high-throughput real time PCR (HUANG et al., 2014) need a previous selection of which strains will be evaluated. Usually, it is expected that bacterial isolates from the same site collection and same species produce the same class of compounds (SCHÄBERLE et al., 2010) and just one is selected for chemical screening. In some cases, even strains from completely different sources produce the same compounds (SONJAK; FRISVAD; GUNDE- CIMERMAN, 2005). The PCA analysis of all extracts revealed that all three strains isolated from the same site collection were different due to their ability of metabolite production. While RKMT_073 produced few natural products in the evaluated growth conditions, RKMT_111 was an exquisite producer of secondary metabolites, including the only one producing the bioactive abyssomicins (in BFM-11m) and RKMT_176 was also a great producer of secondary metabolites (e. g. anthramycin, terragine, talomone, diolmycin, etc.). Among all strains, the PCA and cluster analysis clearly revealed that RKMT_111 and RKMT_176 present the most similar capability of metabolites production since they fell close to each other in the correspondent media. Also, seven compounds were identified as produced by both strains (Figures S7-S13 in Supplementary Material). These results revealed the importance of analyzing carefully all strains isolated in the search of new compounds, a hard task that appears to be more and more difficult every day.

5.2.3. Biotransformations in bacterial isolates

The strain RKMT_070 belonging to the species Eryhtrobacter vulgaris was isolated from a sediment sample and studied for its capability of natural products production. Erythrobacter vulgaris was first reported in 2005 and was isolated from the starfish Stellaster equestris (IVANOVA et al., 2005). The draft of the genome of this species reported interesting enzymes with potential use in industrial processes like beta-galactosidase and esterase-lipase (YAAKOP et al., 2015). An isolated from the genus Erythrobacter is reported as producer of the natural products Erythrazoles A and B, the last one showing cytotoxicity against lung cancer cells (HU; MACMILLAN, 2011). Biotransformation of known compounds by Erythrobacter sp. has not been reported before. 88

In our study we were first searching for compounds produced by the marine isolate in study. However, after purification and identification of the compounds from the broth it was detected that the isolated compounds were new derivatives of substances already present in the cultivation media. This is the first report of 3-acetyl-glycocholic acid and 3-acetyl- glycodeoxycholic acid, both identified in the present study. We propose that these compounds were produced by the acylation of the C3 hydroxyl groups from the respective precursor analogs. Similar biotransformations have been reported before as performed by enzymes isolated from other proteobacteria species, like the regioselective C3 acylation of steroids by an enzyme from Chromobacterium viscosum (RIVA; KLIBANOV, ALEXANDER, 1988). Also, the regioselective acylation of C21 in hydroxysteroids by the enzyme Acetyltransferase I extracted from bacteria has been reported before (MOSA et al., 2015). Such approach for the modifications of known compounds is much better than synthesis, since chemical acylation requires highly polluting compounds like anhydrides, chlorides and pyridine and are frequently not regioselective, leading to more than one analog (MOSA et al., 2015). No other analogs have been detected in the broth of E. vulgaris. In closing, this is the first report of acylation of known metabolites by a Erythrobacter vulgaris marine isolate, an important feature which may be used for industrial and biotechnological applications, especially in the green chemistry field where reduction of harmful solvents is always required. Also it is the first report of the two cholic acid derivatives reported which may present some other properties yet to be explored.

6. Conclusion

The initial main goal of the project was to study the by-catch fauna of shrimp fisheries from the state of São Paulo. Despite the great variety of marine invertebrates captured, its low abundance per species did not allow the isolation of compounds from the species. Among all invertebrates studied, only L. senegalensis showed a potential source of sulphated saponins with reported bioactivity. On the other hand, the assessment of the marine bacteria associated to the invertebrates showed a promising source of known and unknown natural products, besides having other biotechnological potentials as in biotransformations. Regarding the techniques utilized, mass spectrometry once again proved to be a fast and reliable approach when dealing with small amounts of samples like in the study of the starfish L. senegalensis. In the study of the marine bacteria as source of natural products, the selective methods for the isolation of actinobacteria were efficient. Also, the starfish L. senegalensis colletcted as by-catch was a prolific source of actinobacteria. For the study of compounds produced by the bacteria, metabolomics was an important tool for the evaluation of large sets of samples, mainly when it is necessary to prioritize specific conditions among several variables. As a future prospect, all marine bacteria isolated may serve as object of study in many projects with different approaches like cultivation media composition studies and co- cultivation, for the exploration of the full potential of the bacteria obtained.

References

ALTSCHUL, S. F. et al. Basic local alignment search tool. Journal of Molecular Biology, v. 215, p. 403-410, 1990.

ATHALYE, M.; LACEY, J.; GOODFELLOW, M. Selective isolation and enumeration of actinomycetes using rifampicin. Journal of Applied Bacteriology, v. 51, p. 289-297, 1981.

BAKKER-WOUDENBERG, I. A J. M. et al. Antimycobacterial agents differ with respect to their bacteriostatic versus bactericidal activities in relation to time of exposure, mycobacterial growth phase, and their use in combination. Antimicrobial Agents and Chemotherapy, v. 49, n. 6, p. 2387-2398, 2005.

BARBRAUD, C. et al. Fisheries bycatch as an inadvertent human-induced evolutionary mechanism. PLoS One, v. 8, n. 4, 2013. doi:10.1371/journal.pone.0060353.

BARROW, C. J.; SUN, H. H. Spiroquinazoline, a novel substance-P inhibitor with a new carbon skeleton, isolated from Aspergillus flavipes. Journal of Natural Products, v. 57, n. 4, p. 471-476, 1994.

BAYDOUN, E. et al. Microbial transformation of nandrolone with Cunninghamella echinulata and Cunninghamella blakesleeana and evaluation of leishmaniacidal activity of transformed products. Steroids, v. 88, p. 95-100, 2014.

BERLINCK, R. G. S. et al. Challenges and rewards of research in marine natural products chemistry in Brazil. Journal of Natural Products, v. 67, n. 3, p. 510-522, Mar. 2004.

BIHELOVIC, F. et al. Total synthesis and biological evaluation of (−)-atrop–abyssomicin C. Organic & Biomolecular Chemistry, v. 11, p. 5413-5424, 2013.

BISTER, B. et al. Abyssomicin C - a polycyclic antibiotic from a marine Verrucosispora strain as an inhibitor of the p-aminobenzoic acid/tetrahydrofolate biosynthesis pathway. Angewandte Chemie International Edition, v. 43, n. 19, p. 2574-2576, 2004.

BLUNT, J. W. et al. Marine natural products. Natural Products Reports, v. 32, n. 2, p. 116-211, 2015.

BOAVENTURA, M. A. D.; LOPES, R. F. A. P.; TAKAHASHI, J. A. Microorganisms as tools in modern chemistry: the biotransformation of 3-indolylacetonitrile and tryptamine by fungi. Brazilian Journal of Microbiology, v. 35, n. 4, p. 345-347, 2004.

BODE, H. B. et al. Big effects from small changes: possible ways to explore nature’s chemical diversity. ChemBioChem, v. 3, n. 7, p. 619-627, July 2002.

BOLTEN, J. F. Museum Boltenianum sive Catalogus cimeliorum e tribus regnis naturae quae olim collegerat.[London: British Museum (Natural History)], 1906.

BOSE, U. et al. LC-MS-based metabolomics study of marine bacterial secondary metabolite and antibiotic production in Salinispora arenicola. Marine Drugs, v. 13, n. 1, p. 249-266, 2015.

BRANCO, J. O.; FREITAS JÚNIOR, F.; CHRISTOFFERSEN, M. L. Bycatch fauna of seabob shrimp trawl fisheries from Santa Catarina State, southern Brazil. Biota Neotropica, v. 15, n. 2, p. 1-14, 2015.

CATTANI, A. P. et al. Avaliação da ictiofauna da fauna acompanhante da pesca do camarão sete-barbas do município de pontal do Paraná, litoral do Paraná, Brasil. Boletim do Instituto de Pesca, v. 37, n. 2, p. 247-260, 2011.

CHRISTIAN, O. E. et al. Using jasplakinolide to turn on pathways that enable the isolation of new chaetoglobosins from Phomospis asparagi. Journal of Natural Products, v. 68, n. 11, p. 1592-1597, 2005.

CLARCK, A. M.; DOWNEY, M. E. Starfishes of the atlantic. Identification guides 3. London: Chapman & Hall, 1992.

CLUCAS, I. Discards and by-catch in shrimp trawl fisheries (928 FIIU/C928). Rome: FAO, 1997.

COLQUHOUN, J. A. et al. Taxonomy and biotransformation activities of some deep-sea actinomycetes. Extremophiles, v. 2, n. 3, p. 269-277, 1998.

DAI, H. et al. Verrucosispora sediminis sp . nov ., a cyclodipeptide-producing actinomycete from deep-sea sediment. International Journal of Systematic and Evolutionary Microbiology, n. 60, p. 1807-1812, 2010.

DALISAY, D. S. et al. Marine sediment-derived Streptomyces bacteria from British Columbia, Canada are a promising microbiota resource for the discovery of antimicrobial natural products. PLoS One, v. 8, n. 10, 2013. doi: 10.1371/journal.pone.0077078.

DAS, S.; LYLA, P.; KHAN, S. A. Marine microbial diversity and ecology: importance and future perspectives. Current Science, v. 90, n. 10, p. 1325-1335, 2006.

DATTA, D.; TALAPATRA, S. N.; SWARNAKAR, S. Bioactive compounds from marine invertebrates for potential medicines–an overview. International Letters of Natural Sciences, v. 7, p. 42-61, 2015.

DE MARINO, S. et al. Bioactive asterosaponins from the starfish Luidia quinaria and Psilaster cassiope. Isolation and structure characterization by two-dimensional NMR spectroscopy. Journal of Natural Products, v. 66, p. 515-519, 2003.

DEBNATH, M.; PAUL, A. K.; BISEN, P. S. Natural bioactive compounds and biotechnological potential of marine bacteria. Current Pharmaceutical Biotechnology, v. 8, p. 253-260, 2007.

DEMEYER, M. et al. Molecular diversity and body distribution of saponins in the sea star Asterias rubens by mass spectrometry. Comparative Biochemistry and Physiology B: Biochemistry and Molecular Biology, v. 168, n. 1, p. 1-11, 2014.

DONG, G. et al. Chemical constituents and bioactivities of starfish. Chemistry and Biodiversity, v. 8, n. 5, p. 740-791, 2011.

DUNCAN, K. et al. Bioprospecting from marine sediments of New Brunswick, Canada: exploring the relationship between total bacterial diversity and actinobacteria diversity. Marine Drugs, v. 12, n. 2, p. 899-925, 2014.

DWORKIN, M. et al. (Ed.). The prokaryotes: a handbook on the biology of bacteria. 3rd ed. New York: Springer, 2006. Archaea. Bacteria: firmicutes, actinomycetes, v. 3.

EAYRS, S. A guide to bycatch reduction in tropical shrimp-trawl fisheries. Revised edition. Rome: FAO, 2007. 108 p.

EDWARDS, U. et al. Isolation and direct complete nucleotide determination of entire genes. Characterization of a gene coding for 16S ribosomal RNA. Nucleic Acids Research, v. 17, n. 19, p. 7843-7853, 1989.

EL AMRAOUI, B. et al. Antifungal and antibacterial activity of marine microorganisms. Annales Pharmaceutiques Françaises, v. 72, n. 2, p. 107-111, 2014.

ELLEUCH, L. et al. Bioactive secondary metabolites from a new terrestrial Streptomyces sp. TN262. Applied Biochemistry and Biotechnology, v. 162, n. 2, p. 579-593, 2010.

FDHILA, F. et al. DD-diketopiperazines: antibiotics active against Vibrio anguillarum isolated from marine bacteria associated with cultures of Pecten maximus. Journal of Natural Products, v. 66, n. 10, p. 1299-1301, 2003.

FELSENSTEIN, J. Confidence limits on phylogenies: an approach using the bootstrap. Evolution, v. 39, n. 4, p. 783-791, July 1985.

FENICAL, W.; JENSEN, P. R. Developing a new resource for drug discovery: marine actinomycete bacteria. Nature Chemical Biology, v. 2, n. 12, p. 666-673, 2006.

FIEDLER, H. P. et al. Marine actinomycetes as a source of novel secondary metabolites. Antonie van Leeuwenhoek, v. 87, n. 1, p. 37-42, 2005.

FIEDLER, H.-P. et al. Proximicin A, B and C, novel aminofuran antibiotic and anticancer compounds isolated from marine strains of the actinomycete Verrucosispora. The Journal of Antibiotics, v. 61, n. 3, p. 158-163, 2008.

FISCHBACH, M. A; WALSH, C. T. Antibiotics for emerging pathogens. Science, v. 325, n. 5944, p. 1089-1093, Aug. 2009.

FORNER, D. et al. Chemical dereplication of marine actinomycetes by liquid chromatography-high resolution mass spectrometry profiling and statistical analysis. Analytica Chimica Acta, v. 805, p. 70-79, 2013. 93

GAMFELDT, L. et al. Marine biodiversity and ecosystem functioning: what’s known and what's next? Oikos, v. 124, n. 3, p. 252-265, 2015.

GAUTHIER, L. et al. Fungal biotransformation of chlorogenic and caffeic acids by Fusarium graminearum: new insights in the contribution of phenolic acids to resistance to deoxynivalenol accumulation in cereals. International Journal of Food Microbiology, v. 221, n. 856, p. 61-68, 2016.

GHASEMI, Y.; RASOUL-AMINI, S.; FOTOOH-ABADI, E. The biotransformation, biodegradation, and bioremediation of organic compounds by microalgae. Journal of Phycology, v. 47, n. 5, p. 969-980, 2011.

GONTANG, E. A.; FENICAL, W.; JENSEN, P. R. Phylogenetic diversity of gram-positive bacteria cultured from marine sediments. Applied and Environmental Microbiology, v. 73, n. 10, p. 3272-3282, 2007.

GOODFELLOW, M. et al. Verrucosispora maris sp. nov., a novel deep-sea actinomycete isolated from a marine sediment which produces abyssomicins. Antonie van Leeuwenhoek, n. 101, p. 185-193, 2012.

GOODFELLOW, M. et al. Verrucosispora fiedleri sp. nov., an actinomycete isolated from a fjord sediment which synthesizes proximicins. Antonie van Leeuwenhoek, v. 103, p. 493-502, 2013.

GORETTI, M. et al. Biotransformation of electron-poor alkenes by yeasts: asymmetric reduction of (4S)-(+)-carvone by yeast enoate reductases. Enzyme and Microbial Technology, v. 45, n. 67, p. 463-468, 2009.

GRAZIANI, E. I. et al. Phaeochromycins A-E, anti-inflammatory polyketides isolated from the soil actinomycete Streptomyces phaeochromogenes LL-P018. Journal of Natural Products, v. 68, n. 8, p. 1262-1265, 2005.

HALL, M. A.; ALVERSON, D. L.; METUZALS, K. I. By-catch: problems and solutions. Marine Pollution Bulletin, v. 41, n. 16, p. 204-219, Jan. 2000.

HARVEY, A. L.; EDRADA-EBEL, R.; QUINN, R. J. The re-emergence of natural products for drug discovery in the genomics era. Nature Reviews Drug Discovery, v. 14, p. 111-129, Jan. 2015.

HAYAKAWA, M.; NONOMURA, H. Humic acid-vitamin agar, a new medium for the selective isolation of soil actinomycetes. Journal of Fermentation Technology, v. 65, n. 5, p. 501-509, 1987.

HAYAKAWA, M.; YOSHIDA, Y.; IIMURA, Y. Selective isolation of bioactive soil actinomycetes belonging to the Streptomyces violaceusniger phenotypic cluster. Journal of Applied Microbiology, v. 96, n. 5, p. 973-981, 2004.

HAYGOOD, M. G. et al. Microbial symbionts of marine invertebrates: opportunities for microbial biotechnology. Journal of Molecular Microbiology and Biotechnology, v. 1, n. 1, p. 33-43, 1999. 94

HEGAZY, M.-E. F. et al. Microbial biotransformation as a tool for drug development based on natural products from mevalonic acid pathway: a review. Journal of Advanced Research, v. 6, p. 17-33, 2015.

HEIDARY, M.; HABIBI, Z. Microbial transformation of androst-4-ene-3,17-dione by three fungal species Absidia griseolla var. igachii, Circinella muscae and Trichoderma virens. Journal of Molecular Catalysis B: Enzymatic, v. 126, p. 32-36, 2016.

HONG, K. et al. Actinomycetes for marine drug discovery isolated from mangrove soils and plants in China. Marine Drugs, v. 7, n. 1, p. 24-44, 2009.

HOPKINS, D. W.; MacNAUGHTON, S. J.; O’DONNELL, A. G. A dispersion and differential centrifugation technique for representatively sampling microorganisms from soil. Soil Biology and Biochemistry, v. 23, n. 3, p. 217-225, 1991.

HOU, Y. et al. Microbial strain prioritization using metabolomics tools for the discovery of natural products. Analytical Chemistry, v. 84, n. 10, p. 4277-4283, 2012.

HU, Y.; MacMILLAN, J. B. Erythrazoles A-B, cytotoxic benzothiazoles from a marine- derived Erythrobacter sp. Organic letters, v. 13, n. 24, p. 6580-6583, 2011.

HUANG, T. et al. Strain prioritization for natural product discovery by a high- throughput real-time PCR method. Journal of Natural Products, v. 77, p. 2296-2303, 2014.

HUANG, X.; MADAN, A. CAP3: a DNA sequence assembly program. Genome Research, n. 906, p. 868-877, 1999.

HYLEMON, P. B.; HARDER, J. Biotransformation of monoterpenes, bile acids, and other isoprenoids in anaerobic ecosystems. FEMS Microbiology Reviews, v. 22, n. 5, p. 475-488, 1998.

IBEKWE, N.; AMEH, S. Hyphenated techniques in liquid chromatography as current trends in natural products analysis. International Research Journal of Pure and Applied Chemistry, v. 7, n. 3, p. 132-149, 2015.

IJARE, O. B. et al. 1H and 13C NMR characterization and stereochemical assignments of bile acids in aqueous media. Lipids, v. 40, n. 10, p. 1031-1041, 2005.

IÓCA, L. P.; ALLARD, P.-M.; BERLINCK, R. G. S. Thinking big about small beings--the (yet) underdeveloped microbial natural products chemistry in Brazil. Natural Product Reports, v. 31, n. 5, p. 646-675, 2014.

ISHIHARA, K. et al. Stereoselective reduction of α-Keto ester and α-Keto amide with marine actinomycetes, Salinispora strains, as novel biocatalysts. Biochemistry Insights, v. 4, p. 29-33, 2011.

IVANOVA, E. P. et al. Erythrobacter vulgaris sp. nov., a novel organism isolated from the marine invertebrates. Systematic and Applied Microbiology, v. 28, n. 2, p. 123-130, 2005.

JENSEN, P. R. et al. Culturable marine actinomycete diversity from tropical Pacific Ocean sediments. Environmental Microbiology, v. 7, n. 7, p. 1039-1048, 2005.

JENSEN, P. R. et al. Species-specific secondary metabolite production in marine actinomycetes of the genus Salinispora. Applied and Environmental Microbiology, v. 73, n. 4, p. 1146-1152, 2007.

JOMON, K. et al. A new antibiotic, ikarugamycin. The Journal of Antibiotics, v. 25, n. 5, p. 271-280, 1972.

KAKINUMA, K.; KENNETH, L.; RINEHART, J. Tryptophan-dehydrobutyrine diketopiperazine, a metabolite of Streptomyces spectabilis. The Journal of Antibiotics, v. 27, n. 10, p. 733-737, 1974.

KAMEKURA, M. et al. Lysis of halobacteria in bacto-peptone by bile acids. Applied and Environmental Microbiology, v. 54, n. 4, p. 990-995, 1988.

KELLER, S. et al. Abyssomicins G and H and atrop-abyssomicin C from the marine Verrucosispora strain AB-18-032. The Journal of Antibiotics, v. 60, n. 6, p. 391-394, 2007.

KIM, B. Y.; KSHETRIMAYUM, J. D.; GOODFELLOW, M. Detection, selective isolation and characterisation of Dactylosporangium strains from diverse environmental samples. Systematic and Applied Microbiology, v. 34, n. 8, p. 606-616, 2011.

KIM, M. et al. Towards a taxonomic coherence between average nucleotide identity and 16S rRNA gene sequence similarity for species demarcation of prokaryotes. International Journal of Systematic and Evolutionary Microbiology, v. 64, pt. 2, p. 346-351, 2014.

KOHN, K. W.; BONO, V. H.; KANN, H. E. Anthramycin, a new type of DNA-inhibiting antibiotic: reaction with DNA and effect on nucleic acid synthesis in mouse leukemia cells. Biochimica et Biophysica Acta, v. 155, n. 1968, p. 121-129, 1967.

KOLLEROV, V. V. et al. Deoxycholic acid transformations catalyzed by selected filamentous fungi. Steroids, v. 107, p. 20-29, 2015.

KÖNIG, G. M. et al. Natural products from marine organisms and their associated microbes. ChemBioChem, v. 7, n. 2, p. 229-238, 2006.

KWAK, H.; HEGEMAN, A. D.; PARK, S. Seasonal changes in metabolic profiles of galls and leaves of Rhus chinensis using gas chromatography mass spectrometry and liquid chromatography quadrupole time-of-flight mass spectrometry. Journal of Plant Biology, v. 57, n. 2, p. 127-135, 2014.

LAATSCH, H. Antibase 2012 upgrade: the natural compound identifier. Weinheim: Wiley, 2012.

LAURSEN, J. B.; NIELSEN, J. Phenazine natural products: biosynthesis, synthetic analogues, and biological activity. Chemical Reviews, v. 104, n. 3, p. 1663-1685, 2004.

LI, H. et al. Biotransformation of limonene by marine bacteria. Chinese Journal of Analytical Chemistry, v. 34, n. 7, p. 946-950, 2006.

LI, J. et al. Scale-up bioprocess development for production of the antibiotic valinomycin in Escherichia coli based on consistent fed-batch cultivations. Microbial Cell Factories, v. 14, n. 1, 2015. doi:10.1186/s12934-015-0272-y.

LIAO, Z. et al. Verrucosispora lutea sp. nov ., isolated from a mangrove sediment sample. International Journal of Systematic and Evolutionary Microbiology, n. 59, p. 2269-2273, 2009.

LIN, Q. X. H.; LI, L. Verrucosispora wenchangensis sp. nov., isolated from mangrove soil. Antonie van Leeuwenhoek, v. 4, n. 102, p. 1-7, 2012.

LIOU, J.-W. et al. The antimicrobial activity of gramicidin A is associated with hydroxyl radical rormation. PLoS One, v. 10, n. 1, 2015. doi: 10.1371/journal.pone.0117065.

LIU, W.; SHEN, B. Genes for production of the enediyne antitumor antibiotic C-1027 in Streptomyces globisporus are clustered with the cagA gene that encodes the C-1027 apoprotein. Antimicrobial Agents and Chemotherapy, v. 44, n. 2, p. 382-392, 2000.

LOPES, R. da G. et al. Fauna acompanhante da pesca camaroeira no litoral do estado de São Paulo, Brasil. Boletim do Instituto de Pesca, v. 28, n. 2, p. 173-188, 2002.

LU, L. et al. A high-resolution LC-MS-based secondary metabolite fingerprint database of marine bacteria. Scientific Reports, v. 4, 2014. doi: 10.1038/srep06537.

MEHNAZ, S. et al. Characterization of a phenazine and hexanoyl homoserine lactone producing Pseudomonas aurantiaca strain PB-St2, isolated from sugarcane stem. Journal of Microbiology and Biotechnology, v. 19, n. 12, p. 1688-1694, 2009.

MINALE, L. et al. Starfish saponins - XVI. Composition of the steroidal glycoside sulphates from the starfish Luidia maculata. Comparative Biochemistry and Physiology Part B: Comparative Biochemistry, v. 80, n. 1, p. 113-118, 1985.

MINCER, T. J. et al. Widespread and persistent populations of a major new marine actinomycete taxon in ocean sediments. Applied and Environmental Microbiology, v. 68, n. 10, p. 5005-5011, 2002.

MOSA, A. et al. Regioselective acetylation of C21 hydroxysteroids by the bacterial chloramphenicol acetyltransferase I. Chembiochem: an European Journal of Chemical Biology, v. 16, n. 11, p. 1670-1679, 2015.

MUHLING, M.; JOINT, I.; WILLETTS, A. J. The biodiscovery potential of marine bacteria: an investigation of phylogeny and function. Microbial Biotechnology, v. 6, n. 4, p. 361-370, 2013.

NAGAOKA, K. et al. Azinomycins A and B, new antitumor antibiotics. The Journal of Antibiotics, v. 39, n. 11, p. 1527-1532, 1986.

NEMOTO, A. et al. Brasiliquinones A, B and C, new benz[a]anthraquinone antibiotics from Nocardia brasiliensis - I. Producing strain, isolation and biological activities of the antibiotics. The Journal of Antibiotics, v. 50, n. 1, p. 18-21, 1997.

NICOLAOU, K. C.; HARRISON, S. T. Total synthesis of abyssomicin C, atrop-abyssomicin C, and abyssomicin D: implications for natural origins of atrop-abyssomicin C. Journal of the American Chemical Society, v. 129, n. 2, p. 429-440, 2007.

NIEWERTH, D. et al. Antileukemic activity and mechanism of drug resistance to the marine Salinispora tropica proteasome inhibitor salinosporamide A (Marizomib). Molecular Pharmacology, v. 86, n. 1, p. 12-19, May 2014.

O’LEARY, W. M. The fatty acids of bacteria. Bacteriological Reviews, v. 26, n. 4, p. 421-447, 1962.

PANKEY, G. A; SABATH, L. D. Clinical relevance of bacteriostatic versus bactericidal mechanisms of action in the treatment of Gram-positive bacterial infections. Clinical Infectious Diseases, v. 38, n. 6, p. 864-870, 2004.

PLUSKAL, T. et al. MZmine 2: modular framework for processing, visualizing, and analyzing mass spectrometry-based molecular profile data. BMC Bioinformatics, v. 11, 2010. doi: 10.1186/1471-2105-11-395.

POOLE, K. Resistance to β-lactam antibiotics. Cellular and Molecular Life Sciences, v. 61, n. 17, p. 2200-2223, 2004.

POPOV, R. S. et al. Metabolite profiling of polar steroid constituents in the Far Eastern starfish Aphelasterias japonica using LC–ESI MS/MS. Metabolomics, v. 10, n. 6, p. 1152-1168, 2014.

PROCÓPIO, R. E. de L. et al. Antibiotics produced by Streptomyces. The Brazilian Journal of Infectious Diseases, v. 16, n. 5, p. 466-471, 2012.

PYLRO, V. S. et al. Brazilian microbiome project: revealing the unexplored microbial diversity-challenges and prospects. Microbial Ecology, v. 67, n. 2, p. 237-241, 2014.

QIU, D.; RUAN, J.; HUANG, Y. Selective isolation and rapid identification of members of the genus Micromonospora. Applied and Environmental Microbiology, v. 74, n. 17, p. 5593-5597, 2008.

RATEB, M. E. et al. Diverse metabolic profiles of a Streptomyces strain isolated from a hyper-arid environment. Journal of Natural Products, v. 74, n. 9, p. 1965-1971, 2011.

RHEIMS, H. et al. Verrucosispora gifhornensis gen. nov., sp. nov., a new member of the actinobacterial family Micromonosporaceae. International Journal of Systematic Bacteriology, n. 48, p. 1119-1127, 1998.

RIVA, S.; KLIBANOV, A. M. Enzymochemical regioselective oxidation of steroids without oxidoreductases. Journal of the American Chemical Society, v. 110, p. 3291-3295, 1988.

RODRÍGUEZ-GARCÍA, A. et al. Complete genome sequence of “Mycobacterium neoaurum” NRRL B-3805, an androstenedione (AD) producer for industrial biotransformation of sterols. Journal of Biotechnology, v. 224, p. 64-65, 2016.

ROH, H. et al. Genome sequence of the abyssomicin- and proximicin-producing marine actinomycete Verrucosispora maris AB-18-032. Journal of Bacteriology, v. 193, n. 13, p. 3391-3392, 2011.

ROWBOTHAM, T. J.; CROSS, T. Rhodococcus coprophilus sp. nov.: an aerobic nocardioform actinomycete belonging to the “rhodochrous” complex. Journal of General Microbiology, v. 100, n. 1, p. 123-138, May 1977.

SAITOU, N.; NEI, M. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Molecular Biology and Evolution, v. 4, n. 4, p. 406-425, 1987.

SANTOS, J. L. dos. Pesca e estrutura populacional do camarão-branco Litopenaeus schmitti (Burkenroad, 1936) na região marinha e estuarina da baixada santista, São Paulo, Brasil. Boletim do Instituto de Pesca, v. 34, n. 3, p. 375-389, 2008.

SCHÄBERLE, T. F. et al. Marine myxobacteria as a source of antibiotics-comparison of physiology, polyketide-type genes and antibiotic production of three new isolates of Enhygromyxa salina. Marine Drugs, v. 8, n. 9, p. 2466-2479, 2010.

SEIPKE, R. F. Strain-level diversity of secondary metabolism in Streptomyces albus. PLoS One, v. 10, n. 1, 2015. doi: 10.1371/journal.pone.0116457.

SEMÊDO, L. T. et al. Isolation and characterization of actinomycetes from Brazilian tropical soils. Microbiological Research, v. 155, n. 4, p. 291-299, 2001.

SEVERINO-RODRIGUES, E.; GUERRA, D. S. F.; LOPES, R. da G. Carcinofauna acompanhante da pesca dirigida ao camarão-sete-barbas (Xiphopenaeus kroyeri) desembarcada na praia do Perequê, Estado de São Paulo, Brasil. Boletim do Instituto de Pesca, v. 28, n. 1, p. 33-48, 2002.

SEVERINO-RODRIGUES, E.; HEBLING, N. J.; LOPES, R. da G. Biodiversidade no produto da pesca de arrasto-de-fundo dirigida ao lagostim, Metanephrops rubellus (Moreira, 1903), desembarcado no litoral do estado de São Paulo, Brasil. Boletim do Instituto de Pesca, v. 33, n. 2, p. 171-182, 2004.

SHIGEMORI, H. et al. Three new metabolites from the marine yeast Aureobasidium pullulans. Journal of Natural Products, v. 61, n. 5, p. 696-698, 1998.

SHIRAI, M. et al. Terpenoids produced by actinomycetes: isolation, structural elucidation and biosynthesis of new diterpenes, gifhornenolones A and B from Verrucosispora gifhornensis YM28-088. The Journal of Antibiotics, v. 63, n. 5, p. 245-250, 2010.

SOLANKI, R.; KHANNA, M.; LAL, R. Bioactive compounds from marine actinomycetes. Indian Journal of Microbiology, v. 48, n. 4, p. 410-431, 2008.

SONJAK, S.; FRISVAD, J. C.; GUNDE-CIMERMAN, N. Comparison of secondary metabolite production by Penicillium crustosum strains, isolated from Arctic and other various ecological niches. FEMS Microbiology Ecology, v. 53, n. 1, p. 51-60, 2005.

SPELLBERG, B. et al. The epidemic of antibiotic-resistant infections: a call to action for the medical community from the Infectious Diseases Society of America. Clinical Infectious Diseases: an Official Publication of the Infectious Diseases Society of America, v. 46, n. 2, p. 155-164, 2008.

SPONGA, F. et al. Biodiversity and potentials of marine-derived microorganisms. Progress in Industrial Microbiology, v. 35, n. 113, p. 65-69, 1999.

SUPONG, K. et al. Verrucosispora andamanensis sp. nov., isolated from a marine sponge. International Journal of Systematic and Evolutionary Microbiology, v. 63, p. 3970-3974, 2013.

TAMURA, K. et al. MEGA6: molecular evolutionary genetics analysis version 6.0. Molecular Biology and Evolution, v. 30, n. 12, p. 2725-2729, 2013.

VARGHESE, T. A.; JAYASRI, M. A.; SUTHINDHIRAN, K. Marine actinomycetes as potential source for histone deacetylase inhibitors and epigenetic modulation. Letters in Applied Microbiology, v. 61, n. 1, p. 69-76, 2015.

VICENTE, J. et al. Biodiversity of actinomycetes associated with caribbean sponges and their potential for natural product discovery. Marine Biotechnology, v. 15, n. 4, p. 413-424, 2013.

WAGMAN, G. H.; WEINSTEIN, M. J. Antibiotics from Micromonospora. Annual Review of Microbiology, v. 34, n. 1, p. 537-558, 1980.

WAGNER, H.; BLADT, S. Plant drug analysis: a thin layer chromatography atlas. 2nd ed. Berlin: Springer, 1996.

WANG, Q. et al. Abyssomicins from the south China sea deep-sea sediment Verrucosispora sp.: natural thioether michael addition adducts as antitubercular prodrugs. Angewandte Chemie International Edition, v. 52, n. 4, p. 1231-1234, 2013.

WARD, J. H. Hierarchical grouping to optimize an objective function. Journal of the American Statistical Association, v. 58, n. 301, p. 236-244, 2011.

WATTANA-AMORN, P. et al. Antibacterial activity of cyclo(L-Pro-L-Tyr) and cyclo(D-Pro- L-Tyr) from Streptomyces sp. strain 22-4 against phytopathogenic bacteria. Natural Product Research, v. 6419, 2015. doi: 10.1080/14786419.2015.1095747.

WILLIAMS, S. T.; DAVIES, F. L. Use of antibiotics for selective isolation and enumeration of actinomycetes in soil. Journal of General Microbiology, v. 38, p. 251-261, 1965.

WYCHE, T. P. et al. First natural analogs of the cytotoxic thiodepsipeptide thiocoraline a from a marine Verrucosispora sp. Journal of Organic Chemistry, v. 76, n. 16, p. 6542-6547, 2011.

100

XI, L. et al. Description of Verrucosispora qiuiae sp. nov., isolated from mangrove swamp sediment and emended description of the genus Verrucosispora. International Journal of Systematic and Evolutionary Microbiology, n. 62, p. 1564-1569, 2012.

XIAO, N. et al. A preliminary phylogenetic analysis of Luidia (Paxillosida: Luidiidae) from Chinese waters with cytochrome oxidase subunit I (COI) sequences. Journal of Ocean University of China, v. 12, n. 3, p. 459-468, 2013.

YAAKOP, A. S. et al. Draft genome sequence of Erythrobacter vulgaris strain O1, a glycosyl hydrolase-producing bacterium. Genomes Announcements, v. 3, n. 3, p. 1-2, 2015.

YANG, S. W. et al. Structure elucidation of a new diketopiperazine Sch 725418 from Micromonospora sp. The Journal of Antibiotics, v. 57, n. 5, p. 345-347, 2004.

YANG, X. W. et al. Isolation, characterization, and bioactivity evaluation of 3-((6-methylpyrazin-2-yl)methyl)-1H-indole, a new alkaloid from a deep-sea-derived actinomycete Serinicoccus profundi sp. nov. Marine Drugs, v. 11, n. 1, p. 33-39, 2013.

YI-LEI, N. et al. Compounds from marine-derived Verrucosispora sp. FIM06054 and their potential antitumour activities. Natural Product Research, v. 28, n. 23, p. 2134-2139, 2014.

101

Supplementary Material

Chromatographic profiles of extracts from the cultivated strains that showed antimicrobial activity. The delay of 0.16 min observed between PDA and ELSD/MS detectors is due to first detection by PDA detector followed by detection by ELSD and MS detectors.

RKMT_071

RT: 0.00 - 10.01 SM: 7B 5.40 NL: 3.98E2 ELS 0.00 A/D Card Ch. 1 A/D 4.68 card RKMT_071 0.00 0 4.70 6.83 NL: 3.20E6 1128.67 TIC MS RKMT_071 100 Full 3.05 3.55 479.29 5.04 7.53 8.46 8.79 560.32 736.96 479.29 1142.69 876.81 876.81 0 4.54 NL: 8.06E4 0.00 Total Scan PDA PDA RKMT_071

uAU 0 4.70 NL: 9.72E5 479.29 m/z= 478.79-479.79 100 MS RKMT_071

0 6.83 NL: 1.88E5 1111.64 m/z= 100 1111.14-1112.14 MS RKMT_071 0 7.52 NL: 4.44E4 1147.64 m/z= 100 4.41 1147.14-1148.14 1147.68 MS RKMT_071 0 3.07 NL: 7.04E4 560.32 m/z= 559.82-560.82 100 MS RKMT_071

0 0 1 2 3 4 5 6 7 8 9 10 Time (min) Figure S1. Chromatographic profile of the extract from Streptomyces sp. strain RKMT_071. ELSD, HRMS and PDA detectors. Below, extracted chromatograms of the ions of m/z 479.2908 [M+H]+, 1111.6445 [M+H]+, 1147.6410 [M+Na]+ and 560.3185 [M+H]+.

RKMT_071 #5440 RT: 4.53 AV: 1 NL: 4.47E5 microAU

227.00000 100

60 327.00000

402.00000 429.00000 0 220 240 260 280 300 320 340 360 380 400 420 440 wavelength (nm) Figure S2. UV profile of the peak at retention time of 4.54 min in the PDA detector from the analysis of the extract from RKMT_071 strain 102

RKMT_160

RT: 0.00 - 10.01 SM: 7B 0.49 ELS NL: 4.96 6.17E1 0.00 0.69 1.75 2.03 3.27 3.44 4.27 4.91 5.64 5.91 6.17 7.07 7.39 8.23 9.03 9.28 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 A/D Card Ch. 50 0.00 0.00 0.00 0.00 1 A/D card RKMT_160 milliVolts 0 NL: 5.71 6.11 6.47 7.05 7.50 7.94 8.39 8.61 4.96 5.40 1.71E4 4.51 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 PDA 100 4.07 0.00 0.00 Total Scan 3.18 3.63 0.00 8.91 0.00 0.00 PDA 50 0.00 0.00 RKMT_160 0 1.84 2.06 NL: 211.14 Full 9.61E5 100 1.47 245.13 3.74 2.91 4.45 TIC MS 0.70 227.14 357.28 5.63 6.53 6.96 8.02 8.55 9.72 RKMT_160 50 301.14 408.31 301.14 301.14 301.14 301.14 301.14 301.14 301.14

0 1.77 NL: 261.12 1.25E5 100 m/z= 260.62- 50 261.62 MS RKMT_160 0 2.19 NL: 284.14 1.39E5 100 m/z= 283.64- 50 284.64 MS RKMT_160 0 0 1 2 3 4 5 6 7 8 9 10 Time (min) Figure S3. Chromatographic profile of the extract from Micromonospora sp. strain RKMT_160. ELSD, PDA and HRMS detectors. Below, extracted chromatograms of the ions of m/z 261.1234 [M+H]+ and 284.1394 [M+H]+

RKMT_178

RT: 0.00 - 10.01 SM: 7B 0.48 ELS NL: 4.07 5.93E1 0.00 1.30 2.26 2.59 3.43 3.82 5.08 5.42 5.93 6.40 6.96 7.73 8.17 9.00 9.13 50 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 A/D Card Ch. 1 A/D card RKMT_178 milliVolts 0 NL: 5.63 5.85 6.47 6.96 7.50 7.94 8.39 8.61 4.96 5.40 1.64E4 4.51 0.00 0.00 0.00 0.00 0.00 0.00 0.00 PDA 100 0.36 4.07 0.00 0.00 0.00 Total Scan 0.00 0.00 0.00 PDA 50 RKMT_178 0 1.84 2.08 3.20 NL: 1.08E6 1.51 211.14 245.13 426.32 3.74 Full 100 0.93 227.14 357.28 4.46 5.07 6.03 6.91 7.24 8.20 8.53 9.79 TIC MS 217.11 301.14 217.11 217.11 217.11 217.11 217.11 217.11 217.11 RKMT_178 50

0 2.52 NL: 345.18 1.35E4 100 m/z= 344.68- 50 345.68 MS RKMT_178 0 2.54 NL: 389.19 5.97E3 100 m/z= 388.25- 50 389.25 MS RKMT_178 0 0 1 2 3 4 5 6 7 8 9 10 Time (min) Figure S4. Chromatographic profile of the extract from Bacillus sp. strain RKMT_178. ELSD, PDA and HRMS detectors. Below, extracted chromatograms of the ions of m/z 345.1842 [M+H]+ and 389.1911 [M+H]+

103

RKMT_184

RT: 0.00 - 10.01 SM: 7B 5.10 NL: 0.00 5.57 2.59E2 0.00 A/D Card ELS Ch. 1 A/D 200 card RKMT_184

100 milliVolts

0 2.12 NL: 0.00 PDA 2.21E4 0.36 Total Scan 20000 0.00 PDA 1.96 8.52 8.61 RKMT_184 15000 5.54 5.76 6.83 7.23 7.72 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

uAU 10000 5000 0 1.78 NL: 211.14 2.07 3.19 Full 1.28E6 100 245.13 426.32 TIC MS 3.73 8.23 8.41 8.74 80 1.47 7.42 RKMT_184 357.28 4.75 6.89 876.81 876.81 876.81 227.14 876.81 876.81 60 339.25 40 20 0 0 1 2 3 4 5 6 7 8 9 10 Time (min) Figure S5. Chromatographic profile of the extract from Halobacillus sp. strain RKMT_184. ELSD, PDA and HRMS detectors

RT: 0.00 - 10.01 SM: 7B 5.10 5.57 NL: 2.59E2 0.00 0.00 A/D Card Ch. 1 A/D card ELS RKMT_184 0 1.78 2.07 3.19 NL: 1.28E6 Full MS1.47 3.73 4.75 7.42 8.05 8.41 211.14 245.13 426.32 6.66 TIC MS RKMT_184 100 227.14 357.28 339.25 876.81 876.81 876.81 876.81 0 1.97 NL: 6.11E3 316.13 Base Peak m/z= 100 316.12-316.14 MS 0 RKMT_184 2.07 351.12 NL: 9.65E3 100 Base Peak m/z= 0 351.11-351.13 MS 2.07 RKMT_184 325.12 100 NL: 9.74E3 Base Peak m/z= 0 2.13 325.11-325.14 MS 217.10 RKMT_184 100 NL: 7.71E4 0 Base Peak m/z= 2.28 349.18 217.08-217.11 MS 100 RKMT_184 0 NL: 1.86E4 2.26 Base Peak m/z= 375.18 100 349.16-349.19 MS RKMT_184 0 NL: 1.08E4 0 1 2 3 4 5 6 7 8 9 10 Base Peak m/z= Time (min) 375.16-375.19 MS RKMT_184 Figure S6.Chromatographic profile of the extract from Halobacillus sp. strain RKMT_184. ELSD and HRMS detectors. Below, extracted chromatograms of the ions of m/z 316.1292 [M+H]+, 351.1213 [M+H]+, 325.1252 [M+H]+, 217.0973 [M+H]+, 349.1791 [M+H]+ and 375.1755 [M+H]+

104

Single Ion Monitoring (SIM) of compounds identified produced by the bacterial isolates studied.

Methyl 3-methoxy-5-methyl-naphthalene-1-carboxylate

RT: 0.0 - 10.0 SM: 15B 2.8 NL: 231.1 5.84E4 100 m/z= 230.6- 80 231.6 MS rkmt_111_b fm-11m_01 60

0 2.8 NL: 231.1 1.89E5 100 m/z= 230.6- 80 231.6 MS rkmt_176_b fm-11m_01 60

0 0 1 2 3 4 5 6 7 8 9 10 Time (min) Figure S7. Extracted chromatograms in SIM mode for the ion [M+H]+ of m/z 231.1016 from RKMT_111 (top) and RKMT_176 (bottom) in BFM-11m corresponding to the compound Methyl 3- methoxy-5-methyl-naphthalene-1-carboxylate

Kurasoin A

RT: 0.0 - 10.0 SM: 15B 3.0 NL: 257.1 2.64E4 100 m/z= 256.6-257.6 A MS 50 rkmt_111_bfm -1m_01 0 3.0 NL: 257.1 1.33E5 100 m/z= 256.6-257.6 B MS 50 RKMT_176_B FM-1m_01 0 3.0 NL: 257.1 4.69E4 100 m/z= 2.6 C 256.6-257.6 257.1 MS 50 rkmt_111_bfm -11m_01 0 3.0 NL: 257.1 2.62E5 100 m/z= D 256.6-257.6 MS 50 rkmt_176_bfm -11m_01 0 0 1 2 3 4 5 6 7 8 9 10 Time (min) Figure S8. Extracted chromatograms in SIM mode for the ion [M+H]+ of m/z 257.1172 corresponding to the compound kurasoin A. A – RKMT_111 in BFM-1m; B – RKMT_176 in BFM-1m; C – RKMT_111 in BFM-11m; D – RKMT_176 in BFM-11m.

105

Anthramycin

RT: 0.0 - 10.0 SM: 7B 2.2 NL: 247.1 4.54E4 100 m/z= 246.5- 80 247.5 MS rkmt_111_b fm-11m_01 60

0 2.2 NL: 247.1 2.24E4 100 m/z= 246.5- 80 247.5 MS rkmt_176_b fm-11m_01 60

0 0 1 2 3 4 5 6 7 8 9 10 Time (min) Figure S9. Extracted chromatograms in SIM mode for the ion [M+H]+ of m/z 247.1082 from RKMT_111 (top) and RKMT_176 (bottom) in BFM-11m corresponding to the compound anthramycin

Talomone

RT: 0.0 - 10.0 SM: 15B 2.5 NL: 223.1 1.69E4 100 m/z= 222.6- 80 223.6 MS rkmt_111_b fm-11m_01 60

0 2.5 NL: 223.1 1.38E4 100 m/z= 222.6- 80 223.6 MS rkmt_176_b fm-11m_01 60

0 0 1 2 3 4 5 6 7 8 9 10 Time (min) Figure S10. Extracted chromatograms in SIM mode for the ion [M+H]+ of m/z 223.0964 from RKMT_111 (top) and RKMT_176 (bottom) in BFM-11m corresponding to the compound talomone

106

4H-1,3-Benzodioxin-4-one, 2,2-dimethyl-5-(2-propen-1-yl)

RT: 0.0 - 10.0 SM: 15B 2.5 NL: 219.1 7.17E4 100 m/z= 218.6- 80 219.6 MS rkmt_111_b fm-11m_01 60

0 2.5 NL: 219.1 5.82E4 100 m/z= 218.6- 80 219.6 MS rkmt_176_b fm-11m_01 60

0 0 1 2 3 4 5 6 7 8 9 10 Time (min) Figure S11. Extracted chromatograms in SIM mode for the ion [M+H]+ of m/z 219.1016 from RKMT_111 (top) and RKMT_176 (bottom) in BFM-11m corresponding to the compound 4H-1,3- Benzodioxin-4-one, 2,2-dimethyl-5-(2-propen-1-yl)-

4H-1-Benzopyran-4-one, 2-butyl-8-hydroxy-5,7-dimethoxy-3-methyl

RT: 0.0 - 10.0 SM: 15B 2.6 NL: 293.1 1.61E4 100 m/z= 292.6- 80 293.6 MS rkmt_111_b fm-11m_01 60

0 2.7 NL: 293.1 2.12E4 100 m/z= 292.6- 80 293.6 MS rkmt_176_b fm-11m_01 60

0 0 1 2 3 4 5 6 7 8 9 10 Time (min) Figure S12. Extracted chromatograms in SIM mode for the ion [M+H]+ of m/z 293.1383 from RKMT_111 (top) and RKMT_176 (bottom) in BFM-11m corresponding to the compound 4H-1- Benzopyran-4-one, 2-butyl-8-hydroxy-5,7-dimethoxy-3-methyl-

107

Butyrolactol

RT: 0.0 - 10.0 SM: 15B 4.4 NL: 521.3 1.16E4 100 m/z= 520.8- 80 521.8 MS rkmt_111_b fm-4m_01 60

0 4.4 NL: 521.3 1.22E4 100 m/z= 520.8- 80 521.8 MS RKMT_176 _BFM- 60 4m_01

0 0 1 2 3 4 5 6 7 8 9 10 Time (min) Figure S13. Extracted chromatograms in SIM mode for the ion [M+H]+ of m/z 521.3475 from RKMT_111 (top) and RKMT_176 (bottom) in BFM-4m corresponding to the compound butyrolactol

108

Table S1. Detected ions produced by LC-HRMS from the three strains of Verrucosispora maris isolated and media where they were detected. Compounds identified by comparison to (1) Antibase and (2) Scifinder database R Molecular Theorical Loading # Loading #AntiBase Hits* BFM- BFM- BFM- BFM- BFM- BFM- ASW- Strain t m/z Adduct íon δppm Possible compound ISP3m MB (min) formula mass Individual # Group Total Bacteria Verrucosispora 1m 2m 3m 4m 5m 11m Am 3.72 463.2730 ― ― ― 1 OI ― ― ― ― ― 3.82 373.2737 ― ― ― 1 OI ― ― ― ― ― 3.82 408.3106 ― ― ― 1 OI ― ― ― ― ― 5.71 685.4660 ― ― ― 2 OI ― ― ― ― ― 5.93 533.3816 [M+H]+ 533.3842 OI -4.97 Virustomycin FD-892 C H O 2 4 1 0 RKMT_073 5.93 565.4077 [M+MeOH+H]+ 32 52 6 565.4104 OI -4.78 (1) 6.91 668.5465 [M+NH ]+ ― OI ― ― ― ― ― 4 No matches 2 6.91 673.5013 [M+Na]+ ― OI ― ― ― ― ― 6.97 645.4706 ― ― ― 2 OI ― ― ― ― ― 7.00 666.5313 [M+NH ]+ 666.5303 OI 1.40 4 C H O 2 1 0 0 Putatively new 7.00 671.4865 [M+Na]+ 39 68 7 671.4857 OI 1.14 1.93 257.0896 ― ― ― 6 24 ― ― ― ― ― 1.99 220.0948 ― ― ― OI OI ― ― ― ― ― 2.00 443.2500 ― ― ― 1 OI ― ― ― ― ― 2.09 463.2203 ― ― ― OI OI ― ― ― ― ― 2.15 237.0850 ― ― ― 1 OI ― ― ― ― ― 2.18 206.0813 ― ― ― 13 15 ― ― ― ― ― 2.18 270.0430 ― ― ― OI OI ― ― ― ― ― 2.19 201.1024 13 15 ― ― ― ― ― Possible analogs (two double bonds) 2.19 205.0973 13 15 ― ― ― ― ― 2.19 247.1082 [M+H]+ 247.1077 13 13 2.02 Anthramycin DC81 C H N O 5 3 0 2.19 269.0896 [M+Na]+ 13 14 2 3 269.0897 13 13 -0.17 (1) 2.27 231.0628 ― ― ― 12 10 ― ― ― ― ― 2.31 203.1278 ― ― ― OI OI ― ― ― ― ― 2.38 271.0601 ― ― ― OI OI ― ― ― ― ― 2.40 213.0787 ― ― ― 6 24 ― ― ― ― ― RKMT_111 2.41 345.1259 ― ― ― 6 24 ― ― ― ― ― 2.42 289.1047 ― ― ― 13 13 ― ― ― ― ― 2.42 371.1245 ― ― ― 7 8 ― ― ― ― ― 2.45 241.1434 OI OI ― ― ― ― ― Possible analogs (two double bonds) 2.49 237.1122 13 15 ― ― ― ― ― + 2.49 240.1231 [M+NH4] 240.1230 13 13 0.25 + 2.50 223.0964 [M+H] C12H14O4 223.0965 13 13 -0.36 38 2 0 Talomone (1) 2.50 245.0784 [M+Na]+ 245.0784 12 11 -0.16 2.50 219.1017 [M+H]+ 219.1016 13 0.38 4H-1,3-Benzodioxin- C H O 13 8 1 0 4-one, 2,2-dimethyl- 13 14 3 251.1278 2.50 251.1276 [M+MeOH+H]+ 13 -0.77 5-(2-propen-1-yl) (2) 2.50 483.1327 ― ― ― 13 15 ― ― ― ― ― 2.53 253.1410 [M+H]+ ― OI OI ― ― ― ― ― + No matches 2.54 297.1672 [M-CO2+H] ― OI OI ― ― ― ― ― 2.57 325.1986 ― ― ― OI OI ― ― ― ― ― 2.57 405.1476 ― ― ― 6 OI ― ― ― ― ―

109

2.58 379.1321 ― ― ― 9 6 ― ― ― ― ― 2.58 295.1518 ― ― ― OI 18 ― ― ― ― ― 2.58 281.0687 [M+H]+ ― 15 ― ― ― ― ― No matches 13 2.58 303.0506 [M+Na]+ ― 15 ― ― ― ― ― 2.58 581.2609 ― ― ― 13 15 ― ― ― ― ― 2.59 283.0656 ― ― ― 13 15 ― ― ― ― ― + 2.64 257.1173 [M+H-2H2O] 257.1172 13 15 0.13 4H-1-Benzopyran-4- + 2.64 275.1278 [M+H-H2O] 275.1278 13 12 -0.05 one, 2-butyl-8- + 2.64 293.1383 [M+H] C16H20O5 293.1384 13 13 -0.27 12 1 0 hydroxy-5,7- + 2.64 310.1648 [M+NH4] 310.1649 13 13 -0.25 dimethoxy-3-methyl 2.64 315.1203 [M+Na]+ 315.1203 12 11 0.11 (2) 2.64 241.1222 6 21 ― ― ― ― ― Possible analogs (one double bond) 2.66 439.1397 5 25 ― ― ― ― ― 2.67 255.1050 ― ― ― 6 24 ― ― ― ― ― 2.67 207.1017 ― ― ― 6 24 ― ― ― ― ― 2.71 495.3406 ― ― ― OI OI ― ― ― ― ― 2.72 315.2277 ― ― ― OI OI ― ― ― ― ― 2.72 347.1423 ― ― ― 6 24 ― ― ― ― ― 2.72 473.3586 ― ― ― OI OI ― ― ― ― ― 2.72 355.2599 ― ― ― 6 24 ― ― ― ― ― 2.72 437.3373 [M+H-H O]+ ― OI OI ― ― ― ― ― 2 No matches 2.73 455.3481 [M+H]+ ― OI OI ― ― ― ― ― 2.72 355.1280 [M-CO +H]+ ― OI OI ― ― ― ― ― 2 No matches 2.73 399.0985 [M+H]+ ― 6 24 ― ― ― ― ― 2.78 392.2068 ― ― ― 13 15 ― ― ― ― ― 2.79 213.0911 [M+H-2H O]+ 213.0910 9 4 0.43 2 26 0 0 2.79 249.1121 [M+H]+ 249.1121 13 12 -0.04 C H O Putatively new 2.80 266.1384 [M+NH ]+ 14 16 4 266.1387 13 12 -1.05 4 26 0 0 2.80 271.0940 [M+Na]+ 271.0941 10 7 -0.14 2.80 325.1987 ― ― ― OI OI ― ― ― ― ― 2.80 263.1278 [M+MeOH+H]+ 263.1278 13 12 -0.07 Methyl 3-methoxy-5- C H O 9 1 0 methyl-naphthalene- 14 14 3 231.1016 9 2.80 231.1016 [M+H]+ 5 0.17 1-carboxylate (1) 2.80 347.1490 [M+H]+ 347.1489 15 0.18 2.80 364.1755 [M+NH ]+ 364.1755 OI 0.21 4 C H O 13 15 4 1 Abyssomicin C (1) 2.80 369.1308 [M+Na]+ 19 22 6 369.1309 15 -0.08 2.80 715.2739 [2M+Na]+ 715.2725 15 1.91 2.80 617.2365 ― ― ― 13 15 ― ― ― ― ― 2.81 253.1413 [M+H]+ ― OI OI ― ― ― ― ― No matches 2.81 285.2060 [M+MeOH+H]+ ― OI OI ― ― ― ― ― 2.82 441.1149 ― ― ― 6 24 ― ― ― ― ― 2.83 339.2142 ― ― ― OI OI ― ― ― ― ― 2.84 293.2110 ― ― ― OI OI ― ― ― ― ― 2.85 239.1638 ― ― ― OI OI ― ― ― ― ― 2.87 498.1986 ― ― ― 13 15 ― ― ― ― ―

110

2.88 683.4714 ― ― ― OI OI ― ― ― ― ― 2.90 262.0975 ― ― ― OI 1 ― ― ― ― ― 2.92 379.1209 [M+H]+ ― 13 15 ― ― ― ― ― No matches 2.92 401.1028 [M+Na]+ ― 13 15 ― ― ― ― ― 2.93 483.1876 ― ― ― 13 15 ― ― ― ― ― 2.97 252.1052 ― ― ― 1 OI ― ― ― ― ― 2.98 365.1595 [M+H]+ 365.1595 OI 0.12 + 2.99 382.1860 [M+NH4] C19H24O7 382.1860 13 15 -0.11 17 0 0 Putatively new 2.98 387.1414 [M+Na]+ 387.1414 15 -0.16 3.01 357.1310 ― ― ― 9 4 ― ― ― ― ― 3.02 508.2080 ― ― ― 13 15 ― ― ― ― ― 3.02 362.9861 ― ― ― OI OI ― ― ― ― ― 3.02 255.1015 Possible analogs (one double bond) 6 21 ― ― ― ― ― 3.03 257.1172 [M+H]+ 257.1172 4 -0.19 + C16H16O3 9 12 1 0 Kurasoin A (2) 3.03 239.1067 [M+H-2H2O] 239.1067 4 0.00 3.03 364.9840 ― ― ― OI OI ― ― ― ― ― 3.03 341.1359 ― ― ― 4 OI ― ― ― ― ― 3.04 838.4531 ― ― ― OI OI ― ― ― ― ― 3.03 295.0941 6 21 ― ― ― ― ― Possible analogs (one double bond) 3.04 297.1097 9 4 ― ― ― ― ― 3.05 241.1223 ― ― ― 4 22 ― ― ― ― ― 3.07 367.2091 [M+H]+ ― OI OI ― ― ― ― ― + No matches 3.07 323.1829 [M-CO2+H] ― OI OI ― ― ― ― ― 3.08 281.1723 ― ― ― OI OI ― ― ― ― ― 3.10 510.2349 ― ― ― 13 15 ― ― ― ― ― 3.10 393.2727 [M+H]+ 393.2708 OI OI 4.89 C H N O 1 1 1 Fortimicin-AS (1) 3.11 807.5231 [2M+Na]+ 17 36 4 6 807.5162 OI OI 8.52 3.11 513.1910 ― ― ― OI OI ― ― ― ― ― 3.11 212.0707 ― ― ― 13 13 ― ― ― ― ― 3.12 425.3009 ― ― ― OI OI ― ― ― ― ― 3.21 252.1053 ― ― ― 2 OI ― ― ― ― ― 3.21 627.3757 ― ― ― OI OI ― ― ― ― ― 3.12 785.5406 1 OI ― ― ― ― ― Possible analogs (-CH - difference) 3.25 799.5570 2 OI OI ― ― ― ― ― 3.26 280.1002 ― ― ― OI OI ― ― ― ― ― 3.27 265.1004 ― ― ― OI OI ― ― ― ― ― 3.29 279.1161 ― ― ― 2 OI ― ― ― ― ― 3.29 234.0947 ― ― ― 2 OI ― ― ― ― ― 3.30 641.3914 [2M+Na]+ ― 3 OI ― ― ― ― ― No matches 3.30 332.1899 [M+Na]+ ― OI OI ― ― ― ― ― 3.34 714.3496 [M+NH ]+ 714.3484 15 1.69 4 C H O 13 2 0 0 Putatively New 3.34 719.3048 [M+Na]+ 38 48 12 719.3038 15 1.36 3.31 657.3864 ― ― ― OI OI ― ― ― ― ― 3.34 436.2329 ― ― ― 13 15 ― ― ― ― ―

111

3.34 331.1539 [M+H-H O]+ 331.1540 13 15 -0.16 2 Abyssomicin D or 3.34 349.1646 [M+H]+ C H O 349.1646 11 16 -0.03 21 5 2 19 24 6 Abyssomicin H (1) 3.34 371.1465 [M+Na]+ 371.1465 13 15 0.10 3.34 207.1017 ― ― ― OI OI ― ― ― ― ― 3.35 366.1910 ― ― ― 13 15 ― ― ― ― ― 3.35 393.1712 ― ― ― 13 15 ― ― ― ― ― 3.35 625.4176 ― ― ― OI 15 ― ― ― ― ― 3.35 313.1434 ― ― ― 13 OI ― ― ― ― ― 3.35 255.1049 ― ― ― OI OI ― ― ― ― ― + 3.35 394.2223 [M+H-H2O] ― 15 ― ― ― ― ― No matches 13 3.35 412.1730 [M+H]+ ― 15 ― ― ― ― ― 3.37 379.1750 ― ― ― 13 15 ― ― ― ― ― 3.38 796.4414 OI OI ― ― ― ― ― Possible analogs (-CH - difference) 3.40 782.4628 2 1 OI ― ― ― ― ― 3.44 484.1966 ― ― ― 13 OI ― ― ― ― ― 3.45 212.0707 ― ― ― 6 24 ― ― ― ― ― 3.45 349.0616 [M+H-H O]+ ― 6 24 ― ― ― ― ― 2 No matches 3.45 367.0915 [M+H]+ ― 6 24 ― ― ― ― ― 3.45 506.1787 ― ― ― 13 15 ― ― ― ― ― 3.47 765.5111 ― ― ― OI OI ― ― ― ― ― 3.52 363.0410 [M+Na]+ ― 6 24 ― ― ― ― ― No matches 3.52 341.0590 [M+H]+ ― 4 23 ― ― ― ― ― 3.57 363.1802 ― ― ― 13 15 ― ― ― ― ― 3.64 793.3520 ― ― ― OI 1 ― ― ― ― ― 3.64 393.2246 ― ― ― OI OI ― ― ― ― ― 3.65 540.2231 ― ― ― 13 15 ― ― ― ― ― 3.71 222.0948 ― ― ― 2 OI ― ― ― ― ― 3.73 290.1750 ― ― ― OI OI ― ― ― ― ― 3.74 824.4740 ― ― ― 1 OI ― ― ― ― ― 3.75 399.1778 [M+Na]+ 399.1778 6 21 -0.10 16α- C H O 8 1 0 Hydroxyprednisolone 21 28 6 377.1959 6 3.75 377.1958 [M+H]+ 24 -0.10 (2) 3.77 879.3911 ― ― ― OI 3 ― ― ― ― ― 3.80 401.1860 ― ― ― OI 1 ― ― ― ― ― 3.80 586.2651 [M+NH ]+ 586.2678 15 -4.64 4 C H ClN O 13 1 1 1 Chinikomycin A (1) 3.81 591.2205 [M+Na]+ 31 37 2 6 591.2232 15 -4.70 3.81 353.2686 ― ― ― OI OI ― ― ― ― ― 3.82 771.5609 [M+H]+ ― 1 OI ― ― ― ― ― No matches 3.83 793.5436 [M+Na]+ ― 1 OI ― ― ― ― ― 3.88 424.0420 ― ― ― 8 9 ― ― ― ― ― 3.88 655.3709 ― ― ― 13 15 ― ― ― ― ― 3.90 987.3300 ― ― ― OI 1 ― ― ― ― ― 3.93 449.3261 ― ― ― OI OI ― ― ― ― ― 3.93 489.3188 ― ― ― OI OI ― ― ― ― ―

112

3.96 672.3686 ― ― ― 6 24 ― ― ― ― ― 3.98 293.2109 ― ― ― OI OI ― ― ― ― ― 4.01 505.1833 [M+Na]+ 505.1846 15 -2.58 Hydroxystaurosporine + C28H26N4O4 13 8 4 0 4.01 500.2280 [M+NH4] 500.2292 15 -2.37 (1) 4.05 707.2275 ― ― ― 13 15 ― ― ― ― ― 4.12 589.2566 ― ― ― OI 1 ― ― ― ― ― 4.20 631.7747 ― ― ― OI 1 ― ― ― ― ― 4.20 632.2764 ― ― ― OI 1 ― ― ― ― ― 4.36 521.3475 [M+H]+ 521.3473 OI OI 0.47 + C30H48O7 3 1 0 Butyrolactol (1) 4.36 538.3741 [M+NH4] 538.3738 OI OI 0.56 4.40 681.8018 ― ― ― OI 3 ― ― ― ― ― 4.50 469.2487 ― ― ― OI 1 ― ― ― ― ― 4.55 770.8180 [M+H]+ ― OI 1 ― ― ― ― ― No matches 4.54 792.8597 [M+Na]+ ― OI 1 ― ― ― ― ― 4.54 767.8399 ― ― ― OI 1 ― ― ― ― ― 4.55 768.3407 ― ― ― OI 1 ― ― ― ― ― 4.60 847.3730 ― ― ― OI 1 ― ― ― ― ― 4.61 1670.7495 ― ― ― OI 3 ― ― ― ― ― 4.78 448.2128 [M+Na]+ OI OI ― ― ― ― ― No matches 4.78 426.2309 [M+H]+ OI OI ― ― ― ― ―

4.80 367.2743 OI 1 ― ― ― ― ― Possible analogs (-CH - difference) 4.86 381.2899 2 OI 1 ― ― ― ― ― 4.97 631.4547 ― ― ― OI OI ― ― ― ― ― 5.12 537.3115 ― ― ― OI 1 ― ― ― ― ― 5.28 270.2791 ― ― ― OI OI ― ― ― ― ― 5.49 435.3371 OI 1 ― ― ― ― ― Possible analogs (-CH - difference) 5.54 449.3525 2 OI 1 ― ― ― ― ― 5.74 605.3743 ― ― ― OI 1 ― ― ― ― ― 5.97 279.1591 ― ― ― OI 19 ― ― ― ― ― 6.23 503.3998 ― ― ― OI 1 ― ― ― ― ― 7.00 671.4865 ― ― ― OI OI ― ― ― ― ― 7.75 591.4972 ― ― ― OI OI ― ― ― ― ― 1.77 204.1246 ― ― ― 1 2 ― ― ― ― ― 1.94 337.0896 ― ― ― 4 20 ― ― ― ― ― 2.02 224.1282 ― ― ― 4 20 ― ― ― ― ― 2.03 381.1477 ― ― ― OI OI ― ― ― ― ― 2.13 455.1690 ― ― ― OI OI ― ― ― ― ― 2.18 270.0430 ― ― ― OI OI ― ― ― ― ― RKMT_176 2.19 269.0896 [M+Na]+ 269.0897 6 13 -0.17 Anthramycin DC81 C H N O 5 3 0 2.19 247.1082 [M+H]+ 13 14 2 3 247.1077 6 13 2.02 (1) 2.27 231.0628 6 10 ― ― ― ― ― Possible analogs (three double bonds) 2.28 237.1122 6 17 ― ― ― ― ― 2.34 231.1128 ― ― ― OI OI ― ― ― ― ― + 2.36 479.2864 [M+H] C24H38N4O6 479.2864 OI OI -0.10 1 1 0 Terragine D (1)

113

2.37 501.2685 [M+Na]+ 501.2684 OI OI 0.30 2.38 271.0601 ― ― ― OI OI ― ― ― ― ― 2.42 289.1047 ― ― ― 6 13 ― ― ― ― ― 2.42 371.1245 7 8 ― ― ― ― ― Possible analogs (one -O- difference) 2.43 387.2025 6 17 ― ― ― ― ― + 2.49 240.1231 [M+NH4] 240.1230 6 13 0.25 + 2.49 245.0784 [M+Na] C12H14O4 245.0784 10 11 -0.16 38 2 0 Talomone (1) 2.49 223.0964 [M+H]+ 223.0965 6 13 -0.36 2.50 251.1276 [M+MeOH+H]+ 251.1278 6 13 -0.77 4H-1,3-Benzodioxin- C H O 8 1 0 4-one, 2,2-dimethyl- 13 14 3 219.1016 6 2.50 219.1017 [M+H]+ 13 0.38 5-(2-propen-1-yl) (2) 2.54 297.1672 ― ― ― OI OI ― ― ― ― ― 2.57 325.1986 ― ― ― OI OI ― ― ― ― ― 2.57 225.1098 ― ― ― 6 17 ― ― ― ― ― 2.57 405.1476 ― ― ― OI OI ― ― ― ― ― 2.58 379.1321 ― ― ― 8 6 ― ― ― ― ― 2.58 295.1518 6 18 ― ― ― ― ― Possible analogs (two double bonds) 2.61 299.1254 4 20 ― ― ― ― ― 2.64 315.1203 [M+Na]+ 315.1203 10 11 0.11 4H-1-Benzopyran-4- 2.64 293.1383 [M+H]+ 293.1384 6 13 -0.27 one, 2-butyl-8- C16H20O5 12 1 0 hydroxy-5,7- 310.1649 6 dimethoxy-3-methyl + 2.64 310.1648 [M+NH4] 13 -0.25 (2) 2.64 241.1222 ― ― ― 4 21 ― ― ― ― ― 2.64 215.1067 ― ― ― 6 17 ― ― ― ― ― 2.65 275.1278 ― ― ― 10 12 ― ― ― ― ― 2.65 233.1172 ― ― ― 6 17 ― ― ― ― ― 2.66 223.0933 ― ― ― 4 20 ― ― ― ― ― 2.71 495.3406 [M+Na]+ ― OI OI ― ― ― ― ― No matches 2.72 473.3586 [M+H]+ ― OI OI ― ― ― ― ― 2.72 355.2599 ― ― ― OI OI ― ― ― ― ― 2.72 437.3373 ― ― ― OI OI ― ― ― ― ― 2.74 229.0856 [M+H]+ 229.0859 6 17 -1.33 + C14H12O3 5 0 0 Putatively New 2.74 211.0755 [M+H-H2O] 211.0754 10 14 0.51 2.74 269.0785 ― ― ― 10 14 ― ― ― ― ― 2.76 243.0991 ― ― ― 6 17 ― ― ― ― ― 2.78 203.1068 ― ― ― 10 14 ― ― ― ― ― 2.79 213.0911 ― ― ― 10 4 ― ― ― ― ― 2.79 249.1121 ― ― ― 10 12 ― ― ― ― ― 2.80 325.1987 ― ― ― OI OI ― ― ― ― ― 2.80 263.1278 [M+MeOH+H]+ 263.1278 10 12 -0.07 Methyl 3-methoxy-5- C H O 9 1 0 methyl-naphthalene- 14 14 3 231.1016 9 2.80 231.1016 [M+H]+ 5 0.17 1-carboxylate (1) 2.80 266.1384 [M+NH ]+ 266.1387 10 12 -1.05 4 C H O 26 0 0 Putatively new 2.80 271.0940 [M+Na]+ 14 16 4 271.0941 9 7 -0.14

114

2.81 245.1171 ― ― ― 6 17 ― ― ― ― ― 2.81 331.1143 ― ― ― 6 17 ― ― ― ― ― 2.81 285.2060 ― ― ― OI OI ― ― ― ― ― 2.83 452.3370 ― ― ― OI OI ― ― ― ― ― 2.85 239.1638 ― ― ― OI OI ― ― ― ― ― 2.85 393.2610 ― ― ― OI OI ― ― ― ― ― 2.87 339.1779 ― ― ― OI OI ― ― ― ― ― 2.90 390.3003 ― ― ― OI OI ― ― ― ― ― 2.90 262.0975 ― ― ― 1 1 ― ― ― ― ― + 3.01 317.1382 [M+H-H2O] 317.1384 14 -0.60 3.01 335.1490 [M+H]+ 335.1489 14 0.18 (-)-Tetrodecamycin + C18H22O6 10 15 1 0 3.01 352.1754 [M+NH4] 352.1755 14 -0.16 (1) 3.01 357.1310 [M+Na]+ 357.1309 4 0.29 3.02 255.1015 ― ― ― 4 21 ― ― ― ― ― 3.03 295.0941 ― ― ― 4 21 ― ― ― ― ― 3.03 257.1172 [M+H]+ 257.1172 10 4 -0.19 + C16H16O3 12 1 0 Kurasoin A (2) 3.03 239.1067 [M+H-2H2O] 239.1067 10 4 0.00 3.03 364.9840 ― ― ― OI OI ― ― ― ― ― 3.03 341.1359 ― ― ― OI OI ― ― ― ― ― + 3.04 292.1543 [M+NH4] 292.1543 10 14 -0.20 + 3.04 297.1097 [M+Na] C16H18O4 297.1097 10 4 -0.12 13 2 0 Diolmycin (1) 3.05 275.1277 [M+H]+ 275.1278 4 20 -0.37 3.04 211.1118 ― ― ― 10 14 ― ― ― ― ― 3.05 271.1328 ― ― ― 4 20 ― ― ― ― ― 3.05 241.1223 ― ― ― 3 22 ― ― ― ― ― 3.06 362.2537 [M+NH ]+ 362.2537 OI OI -0.10 4 C H O 1 0 0 Putatively new 3.07 367.2091 [M+Na]+ 18 32 6 367.2091 OI OI -0.07 3.08 281.1723 ― ― ― OI OI ― ― ― ― ― 3.10 271.0941 ― ― ― OI OI ― ― ― ― ― 3.11 212.0707 ― ― ― 6 13 ― ― ― ― ― 3.22 228.1383 [M+H]+ ― 1 2 ― ― ― ― ― + No matches 3.22 245.1649 [M+NH4] ― 1 2 ― ― ― ― ― 3.24 767.4875 ― ― ― OI OI ― ― ― ― ― 3.25 272.0917 ― ― ― OI OI ― ― ― ― ― 3.33 399.1705 ― ― ― 1 2 ― ― ― ― ― 3.34 207.1017 ― ― ― OI OI ― ― ― ― ― 3.35 255.1049 ― ― ― OI OI ― ― ― ― ― 3.36 535.4109 ― ― ― OI OI ― ― ― ― ― 3.40 803.4863 ― ― ― OI OI ― ― ― ― ― 3.52 341.0590 ― ― ― 2 23 ― ― ― ― ― 3.56 810.4578 ― ― ― OI OI ― ― ― ― ― 3.57 795.5212 ― ― ― OI OI ― ― ― ― ― 3.61 308.1392 ― ― ― OI OI ― ― ― ― ― 3.61 280.1443 ― ― ― OI OI ― ― ― ― ―

115

3.62 293.2106 ― ― ― OI OI ― ― ― ― ― 3.64 793.3520 ― ― ― 1 1 ― ― ― ― ― 3.64 393.2246 ― ― ― OI OI ― ― ― ― ― 3.68 355.2378 1 2 ― ― ― ― ― Possible analogs (two double bonds) 3.68 359.2404 OI OI ― ― ― ― ― 3.69 328.1542 ― ― ― 4 20 ― ― ― ― ― 3.75 399.1778 ― ― ― 4 21 ― ― ― ― ― 3.80 401.1860 [M+H]+ ― 1 1 ― ― ― ― ― No matches 3.80 423.1679 [M+Na]+ ― 1 2 ― ― ― ― ― 3.81 353.2686 ― ― ― OI OI ― ― ― ― ― 3.82 408.3106 OI OI ― ― ― ― ― Possible analogs (one -O- difference) 3.88 424.0420 6 9 ― ― ― ― ― 3.90 982.3739 [M+NH ]+ ― 1 2 ― ― ― ― ― 4 No matches 3.90 987.3300 [M+Na]+ ― 1 1 ― ― ― ― ― 3.93 449.3261 ― ― ― OI OI ― ― ― ― ― 3.93 489.3188 ― ― ― OI OI ― ― ― ― ― 4.02 1096.5225 ― ― ― 1 2 ― ― ― ― ― 4.08 313.2273 ― ― ― 1 2 ― ― ― ― ― 4.12 589.2566 1 1 ― ― ― ― ― Possible analogs (two double bonds) 4.20 585.4589 OI OI ― ― ― ― ― 4.20 631.7747 ― ― ― 1 1 ― ― ― ― ― 4.20 632.2764 ― ― ― 1 1 ― ― ― ― ― 4.29 1254.5761 ― ― ― 1 2 ― ― ― ― ― 4.36 521.3475 [M+H]+ 521.3473 OI OI 0.47 + C30H48O7 3 1 0 Butyrolactol (1) 4.36 538.3741 [M+NH4] 538.3738 OI OI 0.56 4.41 467.2330 1 2 ― ― ― ― ― Possible analogs (one double bond) 4.50 469.2487 1 1 ― ― ― ― ― 4.50 363.2428 ― ― ― 1 2 ― ― ― ― ― 4.52 793.3609 ― ― ― 1 2 ― ― ― ― ― 4.55 770.8180 [M+H]+ ― 1 1 ― ― ― ― ― No matches 4.54 792.8597 [M+Na]+ ― 1 1 ― ― ― ― ― 4.54 767.8399 ― ― ― 1 1 ― ― ― ― ― 4.55 768.3407 ― ― ― 1 1 ― ― ― ― ― 4.60 847.3730 ― ― ― 1 1 ― ― ― ― ― 4.66 870.3855 ― ― ― 1 2 ― ― ― ― ― 4.78 448.2128 [M+Na]+ ― OI OI ― ― ― ― ― No matches 4.78 426.2309 [M+H]+ ― OI OI ― ― ― ― ― 4.80 367.2743 1 1 ― ― ― ― ― Possible analogs (-CH - difference) 4.86 381.2899 2 1 1 ― ― ― ― ― 4.92 343.2842 ― ― ― 10 14 ― ― ― ― ― 4.97 631.4547 ― ― ― OI OI ― ― ― ― ― 5.12 537.3115 [M+H]+ 1 1 ― ― ― ― ― No matches 5.14 559.2933 [M+Na]+ 1 2 ― ― ― ― ― 5.15 535.2958 Possible analogs (one double bond) 1 2 ― ― ― ― ―

116

5.36 433.3212 1 2 ― ― ― ― ― Possible analogs (one double bond) 5.49 435.3371 1 1 ― ― ― ― ― 5.54 449.3525 ― ― ― 1 1 ― ― ― ― ― 5.74 605.3743 ― ― ― 1 1 ― ― ― ― ― 5.88 353.2811 ― ― ― 1 2 ― ― ― ― ― 5.97 279.1591 ― ― ― 5 19 ― ― ― ― ― 6.11 501.3841 1 2 ― ― ― ― ― Possible analogs (one double bond) 6.23 503.3998 1 1 ― ― ― ― ― 6.34 517.4156 ― ― ― 1 2 ― ― ― ― ― 6.54 673.4371 ― ― ― 1 2 ― ― ― ― ― 6.91 673.5013 Possible analogs (one double bond) OI OI ― ― ― ― ― 7.00 671.4865 [M+Na]+ 671.4857 OI OI 1.14 + C39H68O7 1 0 0 Putatively New 7.00 666.5313 [M+NH4] 666.5303 OI OI 1.40 7.32 569.4468 ― ― ― 1 2 ― ― ― ― ― 7.75 591.4972 ― ― ― OI OI ― ― ― ― ― 7.96 646.2589 ― ― ― OI OI ― ― ― ― ― 8.06 741.4998 ― ― ― 1 2 ― ― ― ― ― 8.27 654.6037 ― ― ― OI OI ― ― ― ― ― 8.84 484.4514 ― ― ― 1 2 ― ― ― ― ― *Source from the compounds were identified and number of compounds found for each ion; OI = Other ions 117

C:\USERS\...\F070_EtOAc 23/02/2015 14:49:47

RT: 0.00 - 9.46 SM: 15B A RT: 0.00 - 9.80 SM: 15B RT: 0.00 - 10.01 SM: 15B 0.53 2.08 NL: 1.02E2 0.41 NL: 1.90 NL: 2.15 4.06E4 2.10 5.94E5 0.0 0.0 A/D Card Ch. 1 A/D 0.0 211.1 100 4.36 6.25 6.89 9.03 0.0 T o tal S can P D A 100 245.1 4.38 Base Peak MS card F070_EtOAc 4.33 5.67 6.34 7.85 8.61 0.0 0.0 0.0 0.0 F 070_E tOA c 457.3 F 070_E tOA c 0.0 0.0 0.0 0.0 0.0 uAU

milliVolts 0 B 0 0 0.55 NL: 1.10E3 NL: 4.48 NL: 3.01E4 3.65E4 0.0 A/D Card Ch. 1 A/D 0.67 301.1 6.47 7.99 8.61 T o tal S can P D A 100 0.92 4.71 6.74 8.20 Base Peak MS 1000 card 0.42 20000 0.0 0.0 0.0 f070_hp20_h2o 301.1 301.1 301.1 413.3 413.3 f070_hp20_h2o f070_hp20_h2o

uAU 0.0

milliVolts 0 0 0 0.53 C NL: 9.23E2 NL: 2.91 NL: 0.0 A/D Card Ch. 1 A/D 3.05E4 466.3 2.19E5 6.34 7.94 8.61 100 2.68 card T o tal S can P D A Base Peak MS 20000 0.41 0.0 0.0 0.0 f070_hp20_meoh- 516.3 f070_hp20_meoh f070_hp20_meoh- uAU 0.0 h2o -h2o

milliVolts 0 h2o 0 0 2.87 NL: 2.90 NL: D NL: 4.13E2 0.0 3.13E4 466.3 8.47E5 2.60 3.25 A/D Card Ch. 1 A/D 5.09 7.67 8.61 100 3.82 4.00 4.55 T o tal S can P D A 2.63 Base Peak MS 0.0 0.0 20000 2.73 0.0 0.0 0.0 f070_hp20_meoh 492.3 f070_hp20_meoh card 0.0 0.0 516.3 f070_hp20_meoh uAU 0.0 milliVolts 0 E 0 0 0.54 2.08 NL: 8.13E1 2.16 NL: 1.90 NL: 3.75 4.36 0.42 5.01E4 2.10 5.44E5 0.0 0.0 6.25 6.83 9.01 A/D Card Ch. 1 A/D 0.0 211.1 4.38 50000 T o tal S can P D A 100 245.1 Base Peak MS 0.0 0.0 0.0 0.0 0.0 card t070_etoac 0.0 5.58 6.11 8.03 8.61 50 t070_eto ac 457.3 t070_eto ac 0.0 0.0 0.0 0.0 uAU

milliVolts 0 0 0 0.52 NL: 8.44E2 NL: 4.48 NL: 0.0 A/D Card Ch. 1 A/D 3.04E4 0.65 301.1 3.04E4 F 6.74 8.03 8.61 100 3.52 6.01 7.29 8.21 card T o tal S can P D A 301.1 Base Peak MS 20000 0.41 0.0 0.0 0.0 t070_hp20_h2o 413.3 413.3 413.3 301.1 t070_hp20_h2o 500 t070_hp20_h2o

uAU 0.0

milliVolts 0 0 0 0.52 NL: 6.11E2 NL: 2.91 NL: 0.0 A/D Card Ch. 1 A/D 2.98E4 466.3 6.68E5 6.78 8.07 8.61 T o tal S can P D A 100 2.66 Base Peak MS 500 2.63 G card 20000 0.41 2.89 0.0 0.0 0.0 t070_hp20_meo h- t070_hp20_meo h 0.0 t070_hp20_meoh- 516.3 uAU 0.0 0.0 h2o -h2o

milliVolts 0 h2o 0 0 2.88 NL: 2.91 NL: NL: 3.51E2 0.0 3.17E4 466.3 9.01E5 2.62 3.26 H A/D Card Ch. 1 A/D 5.09 5.98 7.94 8.61 100 4.02 T o tal S can P D A 2.66 3.31 Base Peak MS 0.0 0.0 card 20000 3.17 0.0 0.0 0.0 0.0 t070_hp20_meo h t070_hp20_meo h 200 0.0 516.3 450.3 t070_hp20_meoh uAU 0.0 milliVolts 0 0 0 0 2 4 6 8 0 2 4 6 8 0 2 4 6 8 10 Time (min) Time (min) Time (min) Figure S14. Comparison of the composition of all fractions obtained by extraction with ethyl acetate and HP-20 of the broth from cultivation of RKMT_070 in tubes and baffled flasks. Left: ELS detector; Center: PDA detector; Right: HRMS detector. A to D: cultivation in baffled flasks, ethyl-acetate, HP- 20 H2O, HP-20 MeOH/H2O, HP-20 MeOH, respectively. E to H: cultivation in tubes, ethyl-acetate, HP-20 H2O, HP-20 MeOH/H2O, HP-20 MeOH, respectively.

8 1 5 11 7 10

6 12 4 9 3

RT: 0.00 - 10.01 SM: 15B 3.74 NL: 1.86 3.19 357.3 6.82E5 100 211.1 2.06 426.3 Base Peak 1.47 245.1 A MS 50 227.1 MB_blank 0 3.76 NL: 466.4 8.47E2 100 m/z= B 465.5-466.5 50 MS MB_blank 0 3.01 NL: 407.3 3.01E4 100 m/z= 3.74 C 406.5-407.5 50 407.3 MS MB_blank 0 2.90 NL: 409.3 1.86E4 100 3.20 409.3 m/z= D 408.5-409.5 50 MS MB_blank 0 3.31 NL: 450.3 4.98E4 100 m/z= E 449.5-450.5 50 MS MB_blank 0 0 1 2 3 4 5 6 7 8 9 10 Time (min) Figure S16. Compounds detected in the medium blank analysis by LC-HRMS. A: Base Peak chromatogram of the medium blank. Extracted ion chromatograms of the ions B: m/z 466.4; C: m/z 407.3; D: m/z 409.3; E: m/z 450.3

RT: 0.00 - 10.01 SM: 15B 3.74 NL: 357.3 6.82E5 100 1.86 3.19 A Base Peak 80 211.1 426.3 MS 2.06 MB_blank 60 245.1 1.47 40 227.1 20 0 2.92 NL: 0.0 1.70E2 100 B m/z= 80 491.5-492.5 3.05 MS 60 2.89 0.0 MB_blank 4.34 9.19 0.82 1.26 2.30 0.0 3.36 4.89 5.52 6.37 6.71 7.02 8.01 8.95 9.29 40 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 20 0 8.51 NL: 0.0 1.30E2 100 C m/z= 80 2.19 7.45 9.46 507.5-508.5 MS 0.0 0.0 0.0 60 2.24 3.21 3.72 6.58 MB_blank 4.77 5.99 7.42 7.75 8.47 9.07 9.85 0.48 2.15 0.0 0.0 0.0 4.38 4.83 0.0 40 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 20 0 0 1 2 3 4 5 6 7 8 9 10 Time (min) Figure S17. Extraction of the ions of m/z 492.3 (B) and m/z 508.3 (C) from analysis of the medium blank (A), confirming the absence of the isolated compounds in the medium composition

119

Figure S18. 1H-NMR spectrum of Fraction 07 (14.0 T, MeOD, ppm)

Figure S19. HMQC contour map of Fraction 07 (14.0 T, MeOD, ppm)

120

Figure S20. HMBC contour map of Fraction 07 (14.0 T, MeOD, ppm)

Figure S21. 1H-NMR spectrum of Fraction 11 (14.0 T, MeOD, ppm)

121

Figure S22. HMQC contour map of Fraction 11 (14.0 T, MeOD, ppm)

Figure S23. HMBC contour map of Fraction 11 (14.0 T, MeOD, ppm)