Universidade Federal Do Rio Grande Do Norte Instituto Metrópole Digital Programa De Pós-Graduação Em Bioinformática

UNIVERSIDADE FEDERAL DO RIO GRANDE DO NORTE INSTITUTO METRÓPOLE DIGITAL PROGRAMA DE PÓS-GRADUAÇÃO EM BIOINFORMÁTICA

EDEN SILVA E SOUZA

Análise do alvo predito da plumieridina em Cryptococcus neoformans

NATAL - RN 2020

EDEN SILVA E SOUZA

Análise do alvo predito da plumieridina em Cryptococcus neoformans

Defesa de Mestrado apresentada ao Programa de Pós- Graduação em Bioinformática da Universidade Federal do Rio Grande do Norte.

Área de concentração: Bioinformática

Linha de Pesquisa: Biologia de Sistemas

Orientadora: Profa. Dra. Marilene Henning Vainstein

NATAL - RN 2020

Catalogação da Publicação na Fonte. UFRN / Biblioteca Setorial do Centro de Biociências

Souza, Eden Silva e. Análise do alvo predito da plumieridina em Cryptococcus neoformans / Eden Silva e Souza. – Natal, RN, 2020.

57 f.

Orientadora: Profa. Dra. Marilene Henning Vainstein

Dissertação (Mestrado) – Universidade Federal do Rio Grande do Norte. Centro de Biociências. Programa Regional de Pós-Graduação em Bioinformática.

1. Cryptococcus neoformans – Dissertação. 2. Plumieridina – Dissertação. 3. Antifúngico – Dissertação. 4. Bioinformática – Dissertaçao. I. Vainstein, Marilene Hennig. II. Universidade Federal do Rio Grande do Norte. III. Título.

RN/UF/BSE-CB CDU 616.98

EDEN SILVA E SOUZA

Análise do alvo predito da plumieridina em Cryptococcus neoformans Defesa de Mestrado apresentada ao Programa de Pós-Graduação em Bioinformática da Universidade Federal do Rio Grande do Norte.

Área de concentração: Bioinformática Linha de Pesquisa: Biologia de Sistemas Orientadora: Profa. Dra. Marilene Henning Vainstein Natal, 28 de fevereiro de 2020.

]

BANCA EXAMINADORA

______Profa. Dra. Marilene Henning Vainstein (Presidente)

______Prof. Dr. Gustavo Antônio de Souza Universidade Federal do Rio Grande do Norte

______Euzebio Guimarães Barbosa Universidade Federal do Rio Grande do Norte

______Charley Christian Staats Universidade Federal do Rio Grande do Sul

AGRADECIMENTOS

Agradeço à minha orientadora Marilene Vainstein pela oportunidade de participar de um projeto que tanto que me permitiu aprender; e ao meu coorientador, João Paulo por todo suporte e paciência durante estes dois anos.

Ao professor Euzebio por toda ajuda nos experimentos in silico.

Aos meus amigos do BioMe, Thiago, Themi, Felipe, Daniela, Raul, Diego, Karla, Maria

Eduarda, Lucas e Tayrone, que riram e se desesperaram comigo durante o mestrado.

Aos integrantes do CBiot, Vanessa, Nicolau, Rafael, Júlia, Eamin, Ane e Matheus, pela disponibilidade de ajudar e ensinar. Agradecimento especial a Vanessa, Julia e Nicolau pelas contribuições durante a escrita do manuscrito.

Aos meus amigos, Riani, Herbeson e Jadilson, que mesmo de longe me aguentam falar sobre minhas aventuras acadêmicas e riem comigo dos problemas.

À Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) pela bolsa concedida no período de abril de 2018 a fevereiro de 2020.

RESUMO

Criptococose é uma infecção fúngica causada por leveduras de Cryptococcus spp. A infecção inicia-se quando células dessecadas ou esporos são inalados e chegam aos pulmões. Se a doença não for propriamente tratada, a infecção pode evoluir, atingir o sistema nervoso central e resultar em meningite meningocócica e até em óbito. O tratamento da criptococose é realizado em três estágios e faz uso de três drogas: fluconazol, anfotericina B e 5-flucitosina. Embora eficaz, o uso destas drogas pode resultar no surgimento de resistência fúngica, efeito tóxico para os pacientes e fármacos com a flucitosina não são comerciados em todos os países. Desta forma, propõe-se investigar o modo de ação do composto antifúngico plumieridina bem como a identificação do seu alvo molecular em C. neoformans. Para isso, realizou-se uma série de experimentos in vitro e in silico. Inicialmente, frações cromatográficas do extrato aquoso das sementes de Allamanda polyantha foram submetidas a ensaios de atividade antimicrobiana. Submeteu-se a fração com atividade antifúngica a análise de ressonância magnética nuclear de carbono e hidrogênio afim de se identificar compostos presentes na amostra. Atividade antifúngica, avaliada através de ensaios antifúngicos, foi de 0.250 mg/mL e o componente majoritário na fração foi a plumieridina. Através da triagem virtual baseada na similaridade do ligante, quitinase foi identificada como alvo molecular da plumieridina. Modelos tridimensionais das quitinases de C. neoformans foram criados e, através do atracamento molecular, observou-se a interação com resíduos do sítio ativo. Ensaios de inibição da atividade quitinolítica mostraram que a atividade é significativamente reduzida na fração secretada e fração celular solúvel. No entanto, a atividade quitinolítica é pouco reduzida pela presença de plumieridina na fração celular insolúvel, onde são necessárias maiores concentrações do composto. Embora plumieridina seja capaz de inibir a atividade quitinolítica, o composto não parece estar relacionado aos níveis transcricionais das quitinases de C. neoformans, reduzindo apenas os níveis transcricionais do gene CHI22. Observou-se que extratos contendo quitinases de macrófagos de camundongo, Bacillus subitilis e de Tenebrio molitor também são inibidos na presença de plumieridina. O tratamento com plumieridina ainda altera o padrão de distribuição dos quitooligômeros na parece celular: de um padrão polarizado para um padrão difuso pela parede. Os resultados validam a predição da triagem virtual e mostram que a inibição da atividade quitinolítica pela plumieridina resulta em divisão celular incompleta e, consequente, atividade antifúngica. Finalmente, os resultados indicam que a plumieridina inibe quitinase e cause morte de C. neoformans, entrentanto, a inibição também ocorre em outros membros da família GH18, indicando que este é um potencial inibidor de GH18.

Palavras-chave: Cryptococcus neoformans, plumieridina, antifúngico, bioinformática, triagem virtual, alvo de drogas, quitinase, GH18.

ABSTRACT Cryptococcosis is a fungal infection caused by yeast from Cryptococcus spp. The infection starts when desiccated cells or spores are inhaled and reach the lungs. If the disease is not properly treated, the infection can progress, reach the central nervous system and result in meningococcal meningitis and even death. Cryptococcosis treatment is carried out in three stages and uses three drugs: fluconazole, amphotericin B and 5-flucytosine. Although effective, the use of these drugs can result in the emergence of fungal resistance, a toxic effect for patients, and drugs as flucytosine are not commercialized worldwide. Thus, it is proposed to investigate the mode of action of the antifungal compound plumieridine as well as the identification of its molecular target in C. neoformans. For this, a series of experiments were carried out in vitro and in silico. Initially, chromatographic fractions of the aqueous extract of Allamanda polyantha seeds were subjected to antimicrobial activity tests. The fraction with antifungal activity was subjected to nuclear magnetic resonance analysis of carbon and hydrogen in order to identify compounds present in the sample. Antifungal activity, evaluated through antifungal tests, was 0.250 mg/mL and the major component in the fraction was plumieridine. Through virtual screening based on ligand’s similarity, chitinase was identified as the molecular target of plumieridine. Three-dimensional models of chitinases from C. neoformans were created and, through molecular docking, interaction with residues from the active site was observed. Chitinolytic activity inhibition assays showed that the activity is significantly reduced in the secreted fraction and soluble cell fraction. However, chitinolytic activity is little reduced by the presence of plumieridine in the insoluble cell fraction, where higher concentrations of the compound are needed. Although plumieridine is able to inhibit chitinolytic activity, the compound does not appear to be related to the transcriptional levels of chitinases of C. neoformans, reducing only the transcriptional levels of the CHI22 gene. It was observed that crude extracts containing chitinases from mouse macrophages, Bacillus subitilis and Tenebrio molitor are also inhibited in the presence of plumieridine. The treatment with plumieridine still alters the distribution pattern of the chitooligomers in the cell wall: from a polarized pattern to a diffuse pattern through the wall. The results validate virtual screening prediction and show that the inhibition of chitinolytic activity by plumieridine results in incomplete cell division and, consequently, antifungal activity. Finally, the results indicate that plumieridine inhibits chitinase and causes death of C. neoformans, however, the inhibition also occurs in other members of the GH18 family, indicating that this is a potential inhibitor of the GH18 family.

Keywords: Cryptococcus neoformans, plumieridine, antifungal, bioinformatics, virtual screening, drug target, chitinase, GH18.

LISTA DE ABREVIATURAS

MIC Concentração inibitória mínima, do inglês: Minimum inhibitory concentration MC Meningite meningocócica, do inglês: Meningococcal meningitis GlcNAc N-acetilglicosamina GXM Glucoronoxilomanana GalXM Galactoxilomanana RT-qPCR PCR quantitativa em tempo real, do inglês: Real time quantitative PCR AmB Anfotericina B 5-FC 5 Flucitocina FZ Fluconazol GH Glicosil hidrolases, do ingles: Glycoside hydrolase AMCase Quitinase ácida de mamíferos, do inglês Acidic mammalian chitinase VS Triagem virtual, do inglês: Virtual screening EC Número EC na classificação de enzimas, do inglês: Enzyme comission

SUMÁRIO

1 INTRODUÇÃO ...... 10

2 REVISÃO DE LITERATURA ...... 11

2.1 Criptococose ...... 11

2.2 Cryptococcus neoformans ...... 12

2.3 Tratamento da criptococose ...... 15

2.4 Predição de alvos moleculares ...... 17

2.5 Quitinases ...... 18

2.6 Inibidores de quitinases ...... 20

2.7 Allamanda polyantha ...... 21

2.8 Justificativa do estudo ...... 22

3 OBJETIVO ...... 23

3.1 Geral ...... 23

3.2 Específicos ...... 23

4 CONCLUSÕES ...... 24

REFERÊNCIAS ...... 25

CAPÍTULO I ...... 32

1 INTRODUÇÃO As variedades de leveduras que compõe o complexo de espécies Cryptococcus neoformans e Cryptococcus gattii são os agentes etiológicos da criptococose (KWON-CHUNG et al., 2014). Cryoptococcus neoformans é uma espécie ubíqua, comumente encontrada em fezes de pombos, que tende a infeccionar indivíduos imunocomprometidos, como recém- transplantados e portadores do vírus HIV/AIDS (LIN, 2009). As estruturas na superfície envolta das células (membrana plasmática, cápsula polissacarídica e parede celular) destas leveduras garantem sucesso na entrada de células hospedeiras, sem que sejam destruídas por componentes do sistema imunológico, e estão relacionadas a patogenicidade das cepas (AGUSTINHO et al., 2018). Determinantes de virulência, como por exemplo, a proteína antifagocítica (App1) e lacases (Lac1 e Lac2), que estão relacionadas à parede celular, atuam como fatores de virulência e de resistência a antifúngicos (ORNER et al., 2019). Criptococose, a infecção causada por Cryptococcus spp., inicia-se quando propágulos são inalados (VELAGAPUDI et al., 2009). Uma vez nos pulmões, as células podem permanecer latentes, evitando fagocitose por conta de componentes da cápsula. Quando fagocitadas, as células conseguem sobreviver e se multiplicar dentro dos fagócitos; a evasão das células de C. neoformans ocorre sem que o sistema imune seja ativido (ZARAGOZA et al., 2009). Nos pulmões estes fungos desenvolvem o quadro de criptococose pulmonar, que se manifesta na forma de tosse crônica, com presença de sangue, dificuldade respiratória e febre (LAM et al., 2001). Ainda nos macrófagos alveolares, as células de C. neoformans se reproduzem e conseguem evadir sem matar as células de defesa e, consequentemente, gerar resposta imunológica (ALVAREZ; CASADEVALL, 2006). Caso não haja diagnóstico precoce ou tratamento adequado, células de C. neoformans podem migrar para o sistema nervoso central, onde desencadeiam o quadro mais comum da infecção, a meningite meningocócica (COLOMBO; RODRIGUES, 2015). O tratamento recomendado da criptococose faz uso de três medicamentos em um regime de três etapas (MOURAD; PERFECT, 2018). A primeira etapa, chamada de indução, faz uso da anfotericina B (AMB) ou AMB lipossomal (LAMB), em conjunto com 5-flucitosina, esta etapa dura de duas a quatro semanas (PERFECT et al., 2010). A LAMB tem mostrado maior atividade antifúngica do que AMB (CHEN, 2002). A etapa de consolidação faz uso de cerca de 800 mg por dia de fluconazol e dura oito semanas (PAPPAS et al., 2009). Por fim, a etapa de manutenção utiliza 200 mg por dia de fluconazol e pode durar de seis a 12 meses (PERFECT et al., 2010). Embora esse seja o tratamento ideal, adequações são feitas em países onde não se tem acesso a certas drogas, como a 5-FC (MOURAD; PERFECT, 2018).

O tratamento citado é mundialmente utilizado, porém estas drogas apresentam efeitos adversos. A AMB, por exemplo, quando utilizada por longos períodos, pode resultar em hepato e nefrotoxicidade, além de náusea, febre e hipóxia (LANIADO-LABORÍN et al., 2009; OSTROSKY-ZEICHNER et al., 2010). O surgimento de linhagens resistentes também é observado durante o uso de AMB (ELLIS, 2002). A monoterapia com fluconazol ou 5-FC também pode resultar no surgimento de cepas resistentes e até desencadear outras infecções fúngicas, o que torna o tratamento com estas drogas combinadas o mais eficaz (SHEEHAN; HITCHCOCK; SIBLEY, 1999; CHEN; SORRELL, 2007). Além do mais, a administração de AMB é feita por via intravenosa, o que demanda pessoal e ambiente adequado para realização do tratamento (HO et al., 2017). Diante do que foi exposto, drogas mais acessíveis, menos tóxicas e que não resultem em resistência são urgentemente necessárias para o tratamento da criptococose. Neste contexto, Bresciani (2013) identificou no extrato aquoso das sementes de Allamanda polyantha iridoides com atividade anti-Cryptococcus spp. Dentre estes iridoides encontra-se a plumieridina. Sabendo que a plumieridina tem atividade fungicida contra C. neoformans, o presente trabalho objetivou a identificação do alvo molecular da plumieridina no fungo e, consequentemente, melhor entendimento da sua atividade anti-Cryptococcus.

2 REVISÃO DE LITERATURA 2.1 Criptococose Criptococose, uma doença reportada em todo mundo, acomete cerca de 1 milhão de pessoas e causa mais de 600,000 mortes por ano (PARK et al., 2009), sendo que, 80% dos casos estão relacionados a pacientes portadores do HIV (JARVIS et al., 2010). Para exemplificar estes dados, estima-se que em 2014, 278 000 portadores do HIV possuíam anticorpos anti- Cryptococcus, destes casos 223 100 resultaram em meningite meningocócica (CM - cryptococcal meningitis) apenas no ano em questão (RAJASINGHAM et al., 2017). Casos de meningite em indivíduos não portadores do vírus HIV geralmente estão associados a transplantes recentes, por exemplo, estima-se que nos Estados Unidos 15% dos casos de CM ocorre nesta coorte de pacientes (PYRGOS et al., 2013) A infecção causada por Cryptococcus spp. pode variar desde uma infecção cutânea (WANG et al., 2015) até infecções sistêmicas fatais (HAYASHIDA et al., 2017). Leveduras de C. neoformans podem ser encontradas no trato respiratório de indivíduos saudáveis sem causar o quadro de infecção, onde hipotetiza-se que o fungo pode ser comensal, residente passageiro ou causador de lesões no epitélio (RANDHAWA; PAL, 1977). O processo infeccioso inicia-se

12 com a inalação dos propágulos que, por conta do tamanho pequeno (1.5 –3.5 µm), alcançam os macrófagos nos alvéolos pulmonares (GEORGE et al., 2017). Porém, indivíduos portadores do HIV/AIDS, em uso de supressores da imunidade celular ou que sofrem de cirrose hepática, são grupos de risco para CM e criptococcemia, que é o quadro sistêmico da doença (LIN; SHIAU; FANG, 2015). Quando a criptococose pulmonar não é tratada, as leveduras podem atingir o sistema nervoso central, onde desenvolve-se a CM. Por conta dos determinantes de virulências de C. neoformans, estas leveduras tem tropismo ao tecido do sistema nervoso central (CNS) (COELHO; BOCCA; CASADEVALL, 2014). Esta doença frequentemente se manifesta na forma de dores de cabeça, febre e vômito e náusea, se não tratada, os sintomas evoluem para confusão mental, perda de consciência e até coma (MOODLEY et al., 2012). Meningite meningocócica é a causa mais comum da doença em pacientes portadores do HIV e frequentemente é fatal (WILLIAMSON et al., 2017).

2.2 Cryptococcus neoformans Cryptococcus neoformans é uma espécie de levedura encapsulada que forma estrutura vegetativa do tipo basídio. Esta espécie é patogênica para humanos e muito utilizada como modelo para o estudo de fatores de virulência em fungos (HULL; DAVIDSON; HEITMAN, 2002). As variedades de C. neoformans junto com as de C. gatti, formam o complexo de espécies patogênicas dentro de Cryptococcus spp. e podem ser diferenciadas através de ensaios bioquímicos (CHANG et al., 2015) e genéticos (BILLMYRE et al., 2002). C. neoformans é uma espécie cosmopolita e o principal agente causador da criptococose. A presença de C. neoformans é historicamente relacionada a presença de aves, em especial, pombos (EMMONS, 1955). No entanto, estas leveduras foram primeiramente isoladas de uma infecção (BUSSE, 1894) e, posteriormente, no mesmo ano foram encontradas no suco de pêra (SANFELICE, 1894). Em 1895 um caso de úlcera epidérmica foi relacionado a presença de Cryptococcus spp., onde, neste paciente, leveduras de também foram encontradas em órgãos, que veio a óbito (MADA; JAMIL; ALAM, 2019). Desde então a comunidade científica tem realizado esforços a fim de entender os mecanismos da virulência de C. neoformans (ESHER; ZARAGOZA; ALSPAUGH, 2018; CUOMO; RHODES; DESJARDINS, 2018; LIU; PERLIN; XUE, 2012) Estudos moleculares, utilizando técnicas como amplified fragment length polymorphisms (AFLP) (HAGEN et al., 2010) e multilocus sequence typing (MLST) (MEYER et al., 2009), tem proposto diferentes classificações a respeito da diversidade de taxa dentro

Cryptococcus spp. Hagen e colaboradores (2015) classificam, com base na genotipagem utilizando as ténicas MLST e PCR-fingerprinting, o complexo de espécies Cryptococcus gatti- Cryptococcus neoformans em sete espécies propostas. O Quadro 1 apresenta o nome usado antes desta classificação, a posição dos clados gerados com base no sequenciamento multilocus (MLST) e o nome específico proposto, porém ainda há debate em relação a quantidade de clados de Cryptococcus spp. Dados indicam que o tipo molecular de C. neoformans mais abundante na América Latina é o VNI (FIRACATIVE; LIZARAZO; ILLNAIT-ZARAGOZÍ; CASTAÑEDA, 2018). No entanto, um estudo realizado no Brasil mostrou que o tipo molecular mais identificado no Sul país é o VNII (WIRTH; AZEVEDO; GOLDAN, 2018).

Quadro 1. Classifição do complexo de espécies Cryptococcus gattii-Cryptococcus neoformans.

Nome específico antigo Clado MLST Tipo molecular Nome específico proposto Cryptococcus neoformans Clado F, AFLP1 VNI Cryptococcus var. grubii Clado G, VNII neoformans AFLP1A/VNB VNII Clado H, AFLP1B

Cryptococcus neoformans Clado I, AFLP2 VNIV Cryptococcus var. neoformans deneoformans

Cryptococcus neoformans AFLP3 VNIII Híbrido variedade híbrida Cryptococcus neoformans x Cryptococcus deneoformans

Cryptococcus gattii Clado D, AFLP4 VGI Cryptococcus gattii Clado C, AFLP5 VGIII Cryptococcus Clado A, AFLP6 VGII bacillisporus Clado E, AFLP7 VGIV

Clado B, AFLP10 VGIV/VGIIIc Cryptococcus deuterogattii Cryptococcus tetragattii Cryptococcus decagattii

Cryptococcus neoformans AFLP8 NA Híbrido var. Cryptococcus neoformans x deneoformans x Cryptococcus gattii hibrido Cryptococcus AFLP4/VGI gattii

Cryptococcus neoformans AFLP9 NA Híbrido var. grubii x Cryptococcus Cryptococcus gattii híbrido AFLP4/VGI neoformans x Cryptococcus gattii

Cryptococcus neoformans AFLP11 NA Híbrido var. grubii x Cryptococcus Cryptococcus gattii híbrido AFLP6/VGII neoformans x Cryptococcus deuterogattii

Os tipos moleculares NA não foram descritos na literatura até o momento. (Adaptado de Hagen et al., 2015).

Para que um fungo tenha sucesso na patogênese são necessárias certas adaptações a resposta imune do hospedeiro, como tolerância a altas temperaturas e variação de pH, alterações na disponibilidade de nutrientes (ORNER et al., 2019). Além disso, os fatores de virulência contribuem para o quadro infeccioso e resistência a drogas. A melanina é um dos fatores de virulência mais estudados em C. neoformans, sua síntese ocorre a partir de compostos fenólicos,

15 como a epinefrina, que se encontram em altas concentrações no sistema nervoso central e contribuem para o tropismo deste fungo neste tecido (WILLIAMSON, 1994; POLACHECK; HEARING; KWON-CHUNG, 1982) e para defesa contra antifúngicos (ALP, 2010). Além da melanina, a cápsula polissacarídica é uma estrutura amplamente estudada, majoritariamente composta por glucoronoxilomanana (GXM), que compõe cerca de 90% do seu conteúdo, e galactoxilomanana (GalXM), que correspode a cerca de 5%. Dentre os outros componentes da cápsula estão as manoproteínas (1%) e substâncias semelhantes a quitina (RODRIGUES et al., 2008; ZARAGOZA et al., 2009). Em C. neoformans, a cápsula sofre alterações em decorrência de condições ambientais (PIERINI; DOERING, 2001) e mutantes de C. neoformans com defeito na cápsula apresentam células viáveis, porém, avirulentos (CHANG; KWON-CHUNG, 1994), o que reforça o papel da cápsula no processo infeccioso. Os mecanismos que o fungo utiliza para evitar fagocitose e “fuga” do sistema imune também estão relacionados à cápsula polissacarídica (ZARAGOZA, 2019).

2.3 Tratamento da criptococose O tratamento da criptococose é usualmente realizado numa combinação de três drogas: AMB, 5-FC e FZ. A meningite meningocócica, que é a forma mais comum da infecção (WILLIAMSON et al., 2017), é tratada em três etapas. Inicialmente, a terapia de indução leva duas semanas de tratamento com AMB aplicado sozinho ou em conjunto com 5-FC, seguido por 8 semanas de terapia de consolidação com doses elevadas de FZ, e por fim, a terapia de manutenção que é realizada com dose menores de fluconazol e pode durar até 12 meses (PERFECT et al., 2010). Os azóis são moléculas orgânicas cíclicas que são classificadas em duas classes: os imidazóis e os triazóis, esta última inclui o fluconazol (ROEMER; KRYSAN, 2014). Em casos onde não é possível realizar o tratamento citado acima, a Sociedade Americana de Doenças Infecciosas - IDSA (Infectious Diseases Society of America) e a Organização Mundial de Saúde - WHO (World Health Organization) recomendam a monoterapia com fluconazol. A monoterapia dura de 10 a 12 semanas onde são administradas altas doses da droga (1200 mg por dia) (PERFECT et al. 2010, WHO 2011). Anfotericina B foi primeiro identificada como um subproduto da fermentação de Streptomyces nodosus. Este antifúngico tem, na maioria, atividade fungistática, e a relação fungicida/fungistática dependa da dose e pH e sensibilidade do microrganismo (GALLIS et al., 1990). Com o passar dos anos desenvolveu-se a anfotericina B lipossomal (LAmB) que apresenta maior eficácia e menor toxicidade (JARVIS et al., 2017; LEENDERS et al., 1997;

HAMILL et al., 2010). Observou-se que a administração de AmB em conjunto com 5-FC aumenta a sobrevivência em pacientes com criptococose meningocócica em relação ao tratamento apenas com AmB. O mesmo não acontece quando AmB é administrada com fluconazol (DAY et al., 2013). Flucitosina (5-FC) é um análogo fluorado sintético da citosina desenvolvida inicialmente como antitumoral (HEIDELBERGER et al., 1958). O uso deste antifúngico é recomendado em conjunto com AmB, uma vez que a monoterapia apenas com 5-FC pode levar a resistência fúngica em aproximadamente 50% dos casos (HOSPENTHAL; BENNETT, 1998; PERFECT et al., 2010). Atualmente a 5-FC não está disponível em alguns países da África, Ásia e Brasil (AGUIAR et al., 2017; ABASSI et al., 2015). Equinocandinas, os antifúngicos mais recentes na indústria farmacêutica, inibem a síntese de 1,3-β-glucano sintase, que sintetiza um dos principais componentes da parede celular fúngica (KURTZ; DOUGLAS, 1997). Estes antifúngicos são divididos em duas classes de compostos: os lipopeptídeos e as papulacandinas (ONISHI et al., 2000). Caspofungina foi a primeira aquinocandina aprovada pela Administração de Alimentos e Medicamentos dos Estados Unidos da América (FDA– Food and Drug Administration) no tratamento da candidíase e aspergilose. Embora a maioria das equinocandinas é de composição peptídica, terpenos já foram identificados com propriedades semelhantes (LEET et al., 1996). Esta classe de antifúngicos ainda não teve atividade observada contra C. neoformans, o que pode estar associado a uma variação na ligação da glucano sintase nesta espécie (MALIGIE; SELITRENNIKOFF, 2005). No Quadro 2 são apresentadas as principais características das drogas utilizadas no tratamento da criptococose.

Quadro 2. Antifúngicos usados no tratamento da criptococose. Classe do Polienos Análogos de Azóis antifúngico nucleosídeo Antifúngico Anfotericina B Flucitosina (5-FC) Fluconazol (AMB)

Vantagens Eficiência Disponível para Baixo custo, disponibilidade farmacológica e administração oral e e administração oral baixas taxas de alta eficiência resistência farmacológica

Desvantagens Administração Hepatotoxicidade Fungistático e intravenosa e severa, disponibilidade desenvolvimento de necessidade de limitada, custo e resistência hospitalização, resistência (quando nefrotoxicidade administrado em severa, custo e monoterapia) disponibilidade

Mecanismo de Liga-se ao Interfere a síntese de Inibe o citocromo p450, ação ergosterol e inibe a DNA e proteínas consequentemente síntese de síntese de ergosterol e integridade da membrana, resulta membrana em morte celular por dano oxidativo

Adaptado de Santos-Gandelman; Machado-Silva (2019)

2.4 Predição de alvos moleculares O processo de descoberta de novos fármacos é custoso e pode levar até 14 anos e custo aproximado de U$ 800 000 (SONG; LIM; TONG, 2009; LAVECCHIA; DI GIOVANNI, 2013). Porém, avanços nas técnicas de cristalografia e na computação de alta performance permitiram o surgimento da triagem de em alta vazão de alvos e ligantes (High-Throughput screening), por exemplo. Nesta abordagem uma molécula específica, um receptor, por exemplo, é cruzada com um banco de estruturas de ligantes, afim de se obter moléculas com maior probabilidade de interação (PINZI; RASTELLI, 2019). A eficiência de abordagens computacionais na descoberta de alvos moleculares pode ser demonstrada pelo aumento do seu uso na academia e indústria farmacêutica nas pesquisas que envolvem genômica, proteômica, quimio e bioinformática (LAVECCHIA; DI GIOVANNI, 2013; BLEICHER et al., 2003; KALYANARAMAN; BERNACKI; JACOBSON, 2005). Outra abordagem amplamente utilizada na predição de moléculas ativas baseia-se na comparação de estruturas de ligantes, tendo como princípio que estruturas semelhantes podem apresentar atividades biológicas semelhantes (KUBINYI, 2002). Neste escopo, a técnica de fingerprint (assinatura) molecular considera os fragmentos da molécula como bits e os descritos variam na forma de decompor a molécula (JAMES, 2004). A métrica de similaridade mais

18 utilizada se baseia no coeficiente de similaridade de Jaccard (BAJUSZ; RÁCZ; HÉBERGER, 2015). Por exemplo, pharmACOphore é uma ferramenta que se baseia no molecular fingerprint e pode criar alinhamentos múltiplos ou comparações entre duas estruturas (KORB et al., 2010) No contexto de descoberta de alvos, o atracamento (ou docking) molecular é definido como simulação da energia de ligação e modo de ligação entre determinado alvo (proteína) e seu ligante (drogas, por exemplo) (BROOIJMANS; KUNTZ, 2003; GRINTER; ZOU, 2014). O virtual screening (VS) utiliza o docking molecular na triagem de alvos para um ligante o qual não se sabe o modo de interação (CHEN; ZHI, 2001). Para realização de VS baseado no docking molecular, a interação de determinado ligante será simulada com o sítio ativo de todas as proteínas de um banco de dados, por fim, os complexos são classificados de acordo com a energia de ligação (XU; HUANG; ZOU, 2018). Metodologias para esta abordagem estão bem estabelecidas na literatura (VASSEUR; BAUD; VIGOUROUX, 2014; XU; HUANG; ZOU, 2018; CHEN; UNG, 2001; PINZI; RASTELLI, 2019.). Uma vez que o genoma, bem como as proteínas, de Cryptococcus neoformans está disponível nos bancos de dados (GERIK; LAM; LODGE, 2014), a busca virtual de alvos é possível.

2.5 Quitinases A parede celular de fungos é a mais exposta da célula e é majoritariamente constituída pelos biopolímeros quitina e glicanos (LATGÉ, 2007). Estes polímeros contribuem para manutenção da integridade e rigidez da parede celular (BANKS et al., 2005), enquanto os monômeros de quitina estão mais relacionados à arquitetura da cápsula polissacarídica (RODRIGUES et al., 2008). Quitina, um biopolímero de poli-β-1,4-N-acetilglicosamina, é o composto natural mais abundante na natureza depois da celulose e está presente no exoesqueleto de crustáceos e parede celular de fungos (AZUMA et al., 2015; MUZZARELLI, 1977). Este polímero também pode ser encontrado na sua forma desacetilada, chamada de quitosano. Este, forma uma estrutura co- polimérica composta por resíduos de D-glucosamina e N-acetilglucosamina (ARANAZ et al., 2009). A remodelagem de quitina, um processo essencial para a célula fúngica, é mantido por um balanço entre a atividade de quitina sintases (2.4.1.16), responsáveis pela síntese de quitina (RONCERO, 2002), e quitinases (3.2.1.14), responsáveis pela degradação da mesma (HARTL et al., 2012). Enzimas quitinolíticas podem ser classificadas em endoquitinases e exoquitinases. As endoquitinases (EC 3.2.1.14) hidrolisam os polímeros de quitina em pontos aleatórios e seus

19 produtos são oligômeros de N-acetilglicosamina. Por outro lado, as exoquitinases se dividem em quitobiosidades (EC 3.2.1.29) e β-(1,4) N-acetilglucosaminidases (EC 3.2.1.30). A primeira atua na extremidade não reduzida e libera diacetilquitobiose (GlcNAc)2, enquanto a última hidrolisa os oligômeros gerados pelas outras enzimas e seus produtos são monômeros de GlcNAc (HARMAN et al., 1993; SEIDL, 2008; OYELEYE; NORMI, 2018) (Figura 1). Enzimas envolvidas na hidrólise de ligação glicosídica fazem parte do grupo de enzimas glicosil hidrolase (GH) (EC 3.2.1.-) (www.cazy.org). Com base na similaridade de sequência as enzimas com atividade quitinolítica foram agrupadas nas famílias GH18, 19 e 20. As GH18 são amplamente encontradas na natureza, com registros em bactérias, fungos, vírus e animais; as GH19 por sua vez, são majoritariamente encontradas em plantas; e as GH20 são encontradas em humanos e bactérias (HENRISSAT, 1991; OYELEYE; NORMI, 2018).

Figura 1. Padão de clivagem das enzimas com atividade quitinolítica. As subunidades de quitina são mostradas em azul claro, o açúcar terminal reduzido em azul escuro. As linhas tracejadas indicam que o polímero é maior do que o ilustrado. Figura traduzida de Seidl (2008).

As quitinases de fungo pertencem a família GH18, salvo um registro para o fungo Nosema bombycis, que possui quitinase da família GH19 (Han et al., 2016; KARLSSON; STENLID, 2018). As quitinases fúngicas participam de processos como expansão celular e divisão ramificação das hifas, germinação de esporos e formação de septos, ou seja, processos envolvidos a crescimento e desenvolvimento morfológico que necessitam da remodelagem da parede celular e geralmente são encontradas em quantidade (SEIDL, 2008; ADAMS, 2004). Observou-se que quitinases de Saccharomyces cerevisiae são necessárias durante a para divisão celular (KURANDA; ROBBINS, 1991). Porém, apesar do papel essencial das quitinases e importância do balanço quitinase/quitinase sintase, estas duas enzimas tem regulação independente em S. cerevisiae e Candica albicans (SELVAGGINI et al., 2004). A deleção em genes de quitinase pode resultar em falha na maturação da estrutura reprodutiva do tipo asco em S. cerevisiae (GIAEVER et al., 2002). Já em C. albicans, o gene de quitinase CHT2 está envolvido em citocinese, enquanto a deleção do gene CHT3 resulta cadeias de células com divisão incompleta (DÜNKLER; SPECHT; ROBBINS, 1995; DÜNKLER et al., 2005). A essencialidade destas proteínas em espécies de fungos sugere que podem ser bons alvos para drogas antifúngicas (RUSH et al., 2010).

2.6 Inibidores de quitinases Inibição da atividade quitinolítica tem aplicações desde o controle de pragas agrícolas ao tratamento de doenças humanas (NAGPURE; CHOUDHARY; GUPTA, 2014; SAGUEZ; VINCENT; GIORDANENGO, 2008). Um número de compostos naturais tem sido descrito como inibidores de quitinase, sendo que, alosamidina foi o primeiro deles (SAKUDA; ISOGAI; SUZUKI, 1987). Alosamidina foi identificado em cultura de Streptomyces sp., é um pseudotrisacarídeo que mimetiza um dos intermediários da quitina e tem atividade inibitória para quitinases da família GH18 devido a sua ação competitiva em relação ao substrato (SAKUDA; ISOGAI; SUZUKI, 1987; BLATTNER; GERARD; SPINDLER-BARTH, 1997; SAKUDA; SAKURADA, 1998). Desde a descoberta desde inibidor, análogos foram criados (HUANG; SHU; CHENG, 2013; HUANG, 2015) e seu potencial como antifúngico, inseticida e na saúde humana tem sido explorado (HUANG; HUANG, 2019). Moléculas de origem peptídica também são relatadas como inibidores de quitinases, como é o caso da argifina e argadina, que foram isolados de espécies de fungos (OMURA et al., 2000; ARAI et al., 2000). Argifina teve sua estrutura caracterizada como um ciclopentapeptídeo. Este composto mostrou atividade inibitória in vivo em larvas de barata, afetando o processo de montagem do exoesqueleto, onde quitinases são essenciais. Mais

21 recentemente, observou-se que mesmo o monopeptídeo dimetilguanil uréia, que é parte da argifina, apresenta atividade, portanto pode ser sintetizado em larga escala (ANDERSEN, et al. 2008). Argadina, também um ciclopentapeptídeo, mostrou atividade inibitória contra quitinases de uma espécie de mosca (Lucilia cuprina) de forma dependente da dose e mais acentuada que argifina (ARAI et al., 2000). Mamíferos não sintetizam quitina, no entando, apresentam quitinases, as quitinases ácidas de mamíferos (acidic mammalian chitinase - AMCase) e seu papel na resposta de doenças respiratórias foi observado em humanos (BOOT et al., 2001; ZHU et al., 2004). Inibição de AMCase em humanos mostrou redução da resposta inflamatória em pacientes asmáticos (ZHU et al., 2004), isso sugere que estes inibidores podem ser usados no tratamento de infecções fúngicas sem causar danos ao paciente. Estudos fitoquímicos já foram realizados com cerca 20% da flora conhecida (NACZK; SHAHIDI, 2006). Isto reforça o papel de plantas na busca de composto biologicamente ativos.

2.7 Allamanda polyantha Apocynaceae é uma das dez famílias de planta mais diversas com 424 gêneros e 4600 espécies (BHADANE, 2018; ENDRESS; LIEDE‐SCHUMANN; MEVE, 2014). Seus membros podem ser encontrados por todo mundo, no entanto, a maior diversidade desta família é encontrada nos trópicos e subtrópicos. Os estudos em Apocynaceae se concentram na polinização, interação planta-inseto e estudos fitoquímicos (FISHBEIN et al., 2018). Dentre os compostos encontrados nesta família pode-se citar os flavonoides, alcaloides, terpenóides e glicosídeos (BHADANE, 2018) e as atividades biológicas incluem anti-inflamatória, antimicrobiana, citotóxica e antioxidante (HÖFLING et al., 2010; BHADANE, 2018). No Brasil ocorrem 95 gêneros de Apocynaceae, dentre eles Allamanda spp. Este gênero, que é endêmico da América do Sul, é amplamente estudado em relação aos seus metabólitos secundários (SAKANE; SHEPHERD, 1986). As flores de A. cathartica tem atividade anti- inflamatória e antioxidante (Góes, 2011) e também apresentam atividade antibacteriana, antifúngica e hepatoprotetora observada in vitro (OMONHINMIN; IJEOMA; UCHE, 2013). A caracterização fitoquímica desta espécie já foi realizada e dentre os constituintes descritos para o caule, folhas, flores e raízes encontram-se: saponinas, compostos fenólicos, carotenoides (para detalhes ver as revisões PETRICEVICH; ABARCA-VARGAS, 2019; AMIN; HEGD, 2016). Embora essa seja uma espécie muito estudada, pouco se sabe sobre as propriedades farmacológicas das suas sementes.

Dentre os compostos encontrados em A. cathartica, o plumierídeo é um dos encontrados em maior quantidade (VILJOEN; MNCWANGI; VERMAAK, 2012). Plumierídeo é um iridoide com atividade antifúngica observada para os fungos dermatófiticos Epidermophyton ﬂoccosum e Microsporum gypseum sem causar toxicidade em células da linhagem P388 (TIWARI; PANDEY; DUBEY, 2012). Outro estudo também observou que não há efeito citotóxico, mutagênico, tóxico ou hemolíticos no extrato das folhas de A. cathartica (BONOMINI et al., 2017). No entanto, outros iridoides já foram isolados de extratos de Allamanda spp., como plumiericina, isoplumiericina, plumierídeo, e plumierídeo coumarato (PETRICEVICH; ABARCA-VARGAS, 2019). Não há na literatura relatos do isolamento da forma aglicona do plumierídeo, chamada plumieridina, de espécies de Allamanda spp. Allamanda polyantha Müll. Arg., também um membro da família Apocynaceae, tem o porte de arbusto e encontra-se distribuída no Brasil principalmente nas regiões Norte e Nordeste (ABDEL-KADER et al., 1997). Iridoides presentes do extrato aquoso das sementes de A. polyantha foram avaliados e identificados através de ressonância magnética nuclear e a sua atividade anti-Cryptococcus foi avaliada. Plumierídeo, plumieridina, plumiericina, isoplumiericina e protoplumiericina foram identificados no extrato aquoso, porém observou-se atividade antifúngica apenas para plumieridina e plumierídeo (BRESCIANI, 2013). Observou- se também que, dentre os compostos identificados, a plumieridina é o mais ativo contra os fungos patogênicos Cryptococcus gattii e C. neoformans. Até então, este composto não tinha sido identificado no gênero Allamanda, embora a literatura relate sua presença no extrato metanólico de Plumeria obtusa, outra Apocynaceae (SALEEM et al., 2011).

2.8 Justificativa do estudo Sem o tratamento adequado a criptococose pode resultar em óbito (COELHO; CASADEVALL, 2016). O tratamento desta infecção fúngica utiliza anfotericina B, fluconazol e 5-flucitosina (PERFECT et al., 2010). No entanto, o uso prolongado destes medicamentos pode ter efeito tóxico e levar a resistência fúngica. Tendo em vista a necessidade no desenvolvimento de fármacos que melhorem o tratamento da criptococose, Bresciani (2013) buscou identificar iridoides com atividade antifúngica em frações do extrato aquoso das sementes de Allamanda polyantha, onde foram identificados o plumierídeo, plumieridina, protoplumiericina e plumieridcina/isoplumiericina. Estes compostos fazem parte dos iridoides, que são monoterpenos com diversas atividades biológicas descritas.

3 OBJETIVO

3.1 Geral

Identificar, por meio da triagem virtual, alvos moleculares da plumieridina em Cryptococcus neoformans, realizar simulações computacionais que expliquem a atividade antifúngica e realizar experimentos in vitro afim de se validar os resultados de predição do alvo.

3.2 Específicos

• Extração, purificação e ressonância magnética nuclear (RMN) da plumieridina.

• Triagem virtual contra o Protein Data Bank na busca de alvos.

• Simular a interação proteína-ligante utilizando as abordagens de docking e dinâmica molecular.

• Determinar a concentração inibitória mínima da plumieridina em C. neoformans var. grubbii H99.

• Realizar ensaios de inibição da atividade do alvo predito.

• Verificar através de PCR quantitativa em tempo real (RT-qPCR) se os níveis transcricionais dos alvos preditos se alteram na presença da plumieridina.

• Microscopia confocal afim de se avaliar alterações morfológicas na parede celular de C. neoformans.

• Verificar se a plumieridina tem ação inibitória apenas para o alvo predito em C. neoformans ou se esta resposta estaria presente em outros organismos.

4 CONCLUSÕES

Atribuiu-se a atividade antifúngica identificada em uma fração cromatográfica do extrato de sementes de Allamanda polyantha a plumieridina, o composto majoritário na fração, identificado por meio de ressonância magnética nuclear. A abordagem de triagem virtual baseada na similaridade dos ligante, utilizada para identificar quitinase como alvo putativo da plumieridina, se mostrou efetivo uma vez que os resultados in vitro mostram que a presença do composto reduz a atividade da enzima. Ensaios de inibição da atividade quitinolítica mostraram que a plumiridina é capaz de reduzir a atividade enzimática na fração secretada e fração celular solúvel de forma dose-dependente. Entretanto, a fração celular insolúvel sofre pouca redução, indicando que maiores concentrações do composto podem ser necessárias. Por meio da análise de interação proteína-ligante, observou-se que a plumieridina tem potencial inibitório para quitinases da família glicosil hidrolase 18 (GH18), por conta da interação com o motivo catalítico. Esta hipótese foi suportada pela redução da atividade quitinolítica observada em sobrenadante de Bacillus subtilis, Tenebrio molitor e macrófagos de camundongos.

REFERÊNCIAS BAJUSZ, D.; RÁCZ, A.; HÉBERGER, Károly. Why is Tanimoto index an appropriate choice for fingerprint-based similarity calculations?. Journal of cheminformatics, v. 7, n. 1, p. 20, 2015. ABASSI, Mahsa; BOULWARE, David R.; RHEIN, Joshua. Cryptococcal meningitis: diagnosis and management update. Current tropical medicine reports, v. 2, n. 2, p. 90-99, 2015. ADAMS, D. J. Fungal cell wall chitinases and glucanases. Microbiology, v. 150, n. 7, p. 2029-2035, 2004. AGUIAR, P. A. D. F. et al. The epidemiology of cryptococcosis and the characterization of Cryptococcus neoformans isolated in a Brazilian University Hospital. Revista do Instituto de Medicina Tropical de São Paulo, v. 59, 2017. AGUSTINHO, D. P. et al. Peeling the onion: the outer layers of Cryptococcus neoformans. Memórias do Instituto Oswaldo Cruz, v. 113, n. 7, 2018. ALP, S. Melanin and its role on the virulence of Cryptococcus neoformans. Mikrobiyoloji bulteni, v. 44, n. 3, p. 519-526, 2010. ALVAREZ, M.; CASADEVALL, A. Phagosome extrusion and host-cell survival after Cryptococcus neoformans phagocytosis by macrophages. Current Biology, v. 16, n. 21, p. 2161-2165, 2006. AMIN, C.; HEGDE, K. Therapeutic Uses of Allamanda Cathartica Linn. With A Note on Its Phrmacological Actions : a Review. International Journal of Pharma And Chemical Research, 2(4), 227-232. 2016. ANAND, N.; BHARTI, C.; RAJINDER, K. G. Chitinases: in agriculture and human healthcare, Critical Reviews in Biotechnology, 34:3, 215-232, 2014. ANDERSEN, O. A. et al. Structure-based dissection of the natural product cyclopentapeptide chitinase inhibitor argifin. Chemistry & biology, v. 15, n. 3, p. 295-301, 2008. ARANAZ, I. et al. Functional characterization of chitin and chitosan. Current chemical biology, v. 3, n. 2, p. 203-230, 2009. AZUMA, K. et al. Chitin, chitosan, and its derivatives for wound healing: old and new materials. Journal of functional biomaterials, v. 6, n. 1, p. 104-142, 2015. BANKS, I. R. et al. A chitin synthase and its regulator protein are critical for chitosan production and growth of the fungal pathogen Cryptococcus neoformans. Eukaryotic Cell, v. 4, n. 11, p. 1902-1912, 2005. BHADANE, B. S. et al. Ethnopharmacology, phytochemistry, and biotechnological advances of family Apocynaceae: A review. Phytotherapy research, v. 32, n. 7, p. 1181-1210, 2018. BILLMYRE, R. B., CROLL, D., LI, W., MIECZKOWSKI, P., CARTER, D. A., CUOMO, C. A., KRONSTAD, J. W., HEITMAN, J. Highly recombinant VGII Cryptococcus gattii population develops clonal outbreak clusters through both sexual macroevolution and asexual microevolution. MBio, 5(4), e01494-14, 2014. BLATTNER, R.; GERALD, P. SPINDLER-BARTH, M. Synthesis and biological activity of allosamidin and allosamidin derivatives. Pesticide Science 50, 312- 318, 1997. BLEICHER, K. H. et al. A guide to drug discovery: hit and lead generation: beyond high- throughput screening. Nature reviews Drug discovery, v. 2, n. 5, p. 369, 2003. BONOMINI, Tiago J. et al. Neuropharmacological and acute toxicological evaluation of ethanolic extract of Allamanda cathartica L. flowers and plumieride. Regulatory Toxicology and Pharmacology, v. 91, p. 9-19, 2017. BOOT, R. G. et al. Identification of a novel acidic mammalian chitinase distinct from chitotriosidase. Journal of Biological Chemistry, v. 276, n. 9, p. 6770-6778, 2001.

BRESCIANI, F. R. Caracterização da atividade antifúngica de extrato aquoso de sementes. Dissertação (Mestrado em Biologia Celular e Molecular). Universidade Federal do Rio Grande do Sul, Porto Alegre, 2013. BROOIJMANS, N.; KUNTZ, I. D. Molecular recognition and docking algorithms. Annual review of biophysics and biomolecular structure, v. 32, n. 1, p. 335-373, 2003. CHANG, Y. C., LAMICHHANE, A. K., BRADLEY, J., RODGERS, L., NGAMSKULRUNGROJ, P., & KWON-CHUNG, K. J. Differences between Cryptococcus neoformans and Cryptococcus gattii in the molecular mechanisms governing utilization of D- amino acids as the sole nitrogen source. PloS one, 10(7), e0131865, 2015. CHANG, Y. C.; KWON-CHUNG, K. J. Complementation of a capsule-deficient mutation of Cryptococcus neoformans restores its virulence. Molecular and cellular biology, v. 14, n. 7, p. 4912-4919, 1994. CHEN, S. C. A. Cryptococcosis in Australasia and the treatment of cryptococcal and other fungal infections with liposomal amphotericin B. Journal of Antimicrobial Chemotherapy, v. 49, n. suppl_1, p. 57-61, 2002. CHEN, S. C. A; SORRELL, T. C. Antifungal agents. Medical Journal of Australia, v. 187, n. 7, p. 404, 2007. CHEN, Y. Z.; UNG, C. Y. Prediction of potential toxicity and side effect protein targets of a small molecule by a ligand–protein inverse docking approach. Journal of Molecular Graphics and Modelling, v. 20, n. 3, p. 199-218, 2001. CHEN, Y. Z.; ZHI, D. G. Ligand–protein inverse docking and its potential use in the computer search of protein targets of a small molecule. Proteins: Structure, Function, and Bioinformatics, v. 43, n. 2, p. 217-226, 2001. COELHO, C.; BOCCA, A. L.; CASADEVALL, Arturo. The tools for virulence of Cryptococcus neoformans. In: Advances in applied microbiology. Academic Press, 2014. p. 1-41. CUOMO, C. A.; RHODES, J.; DESJARDINS, C. A. Advances in Cryptococcus genomics: insights into the evolution of pathogenesis. Memórias do Instituto Oswaldo Cruz, v. 113, n. 7, 2018. DAY, J. N., CHAU, T. T., WOLBERS, M., MAI, P. P., DUNG, N. T., MAI, N. H., ... & THAI, L. H. Combination antifungal therapy for cryptococcal meningitis. New England Journal of Medicine, 368(14), 1291-1302, 2013. DÜNKLER, A. et al. Candida albicans CHT3 encodes the functional homolog of the Cts1 chitinase of Saccharomyces cerevisiae. Fungal Genetics and Biology, v. 42, n. 11, p. 935- 947, 2005. ELLIS, D. Amphotericin B: spectrum and resistance. Journal of Antimicrobial Chemotherapy, v. 49, n. 1, p. 7-10, 2002. EMMONS, C. W. Saprophytic Sources of Gryptococcus neoformans associated with the Pigeon (Columba livia). American Journal of Hygiene, v. 62, n. 3, p. 227-32, 1955. ENDRESS, M. E.; LIEDE-SCHUMANN, S.; MEVE, U. An updated classification for Apocynaceae. Phytotaxa, v. 159, n. 3, p. 175-194, 2014. ESHER, S. K.; ZARAGOZA, O.; ALSPAUGH, J. A. Cryptococcal pathogenic mechanisms: a dangerous trip from the environment to the brain. Memorias do Instituto Oswaldo Cruz, v. 113, n. 7, 2018. FIRACATIVE, C., LIZARAZO, J., ILLNAIT-ZARAGOZI, M. T., & CASTAÑEDA, E. The status of cryptococcosis in Latin America. Memórias do Instituto Oswaldo Cruz, 113(7), 2018. FISHBEIN, M. et al. Evolution on the backbone: Apocynaceae phylogenomics and new perspectives on growth forms, flowers, and fruits. American Journal of Botany, v. 105, n. 3, p. 495-513, 2018.

GALLIS, H. A., DREW, R. H., & PICKARD, W. W. Amphotericin B: 30 years of clinical experience. Reviews of infectious diseases, 12(2), 308-329, 1990. GEORGE, I. A., SPEC, A., POWDERLY, W. G., & SANTOS, C. A. Comparative epidemiology and outcomes of HIV, non-HIV non-transplant and organ transplant associated cryptococcosis: a population-based study. Clinical Infectious Diseases, 2017. GERIK, K. J.; LAM, W.; LODGE, J. K. Analysis of the genome and transcriptome of Cryptococcus neoformans var. grubii reveals complex RNA expression and microevolution leading to virulence attenuation. PLoS genetics, v. 10, n. 4, 2014. GIAEVER, G. et al. Functional profiling of the Saccharomyces cerevisiae genome. Nature, v. 418, n. 6896, p. 387, 2002. GORMAN, J. A. et al. Ascosteroside, a New Antifungal Agent from Ascotricha amphitricha. The Journal of antibiotics, v. 49, n. 6, p. 547-552, 1996. GRINTER, Sam Z.; ZOU, Xiaoqin. Challenges, applications, and recent advances of protein- ligand docking in structure-based drug design. Molecules, v. 19, n. 7, p. 10150-10176, 2014. HAGEN, F., ILLNAIT-ZARAGOZI, M. T., BARTLETT, K. H., SWINNE, D., GEERTSEN, E., KLAASSEN, C. H., ... & MEIS, J. F. In vitro antifungal susceptibilities and amplified fragment length polymorphism genotyping of a worldwide collection of 350 clinical, veterinary, and environmental Cryptococcus gattii isolates. Antimicrobial agents and chemotherapy, 54(12), 5139-5145, 2010. HAGEN, F., KHAYHAN, K., THEELEN, B., KOLECKA, A., POLACHECK, I., SIONOV, E., FALK, R., PARNMEN, S., LUMBSCH, H. T., BOEKHOUT, T. Recognition of seven species in the Cryptococcus gattii/Cryptococcus neoformans species complex. Fungal Genetics and Biology, 78, 16-48, 2015. HAMILL, R. J., SOBEL, J. D., EL-SADR, W., JOJHNSON, P. C., GRAYBILL, J. R., JAVALY, K., ... & AmBisome Cryptococcal Meningitis Study Groupa. Comparison of 2 doses of liposomal amphotericin B and conventional amphotericin B deoxycholate for treatment of AIDS-associated acute cryptococcal meningitis: a randomized, double-blind clinical trial of efficacy and safety. Clinical infectious diseases, 51(2), 225-232, 2010. HAN, B., ZHOU, K., LI, Z., SUN, B., NI, Q., MENG, X., ... & ZHOU, C. Characterization of the first fungal glycosyl hydrolase family 19 chitinase (NbchiA) from Nosema bombycis (Nb). Journal of Eukaryotic Microbiology, 63(1), 37-45, 2016. HARMAN, G. E. et al. Chitinolytic enzymes of Trichoderma harzianum: purification of chitobiosidase and endochitinase. Phytopathology, v. 83, n. 3, p. 313-318, 1993. HARTL, L., ZACH, S., & SEIDL-SEIBOTH, V. Fungal chitinases: diversity, mechanistic properties and biotechnological potential. Applied microbiology and biotechnology, 93(2), 533-543. 2012. HAYASHIDA, M. Z. et al. Disseminated cryptococcosis with skin lesions: report of a case series. Anais brasileiros de dermatologia, v. 92, n. 5, p. 69-72, 2017. HEIDELBERGER, C., GRIESBACH, L., MONTAG, B. J., MOOREN, D., CRUZ, O., SCHNITZER, R. J., & GRUNBERG, E. Studies on fluorinated pyrimidines: II. effects on transplanted tumors. Cancer research, 18(3), 305-317, 1958. HENRISSAT, Bernard. A classification of glycosyl hydrolases based on amino acid sequence similarities. Biochemical journal, v. 280, n. 2, p. 309-316, 1991. HO, J. et al. Intrathecal amphotericin B: a 60-year experience in treating coccidioidal meningitis. Clinical Infectious Diseases, v. 64, n. 4, p. 519-524, 2017. HÖFLING, J. F. et al. Antimicrobial potential of some plant extracts against Candida species. Brazilian Journal of Biology, v. 70, n. 4, p. 1065-1068, 2010. HOSPENTHAL, D. R., & BENNETT, J. E. Flucytosine monotherapy for cryptococcosis. Clinical infectious diseases, 27(2), 260-264. 1998.

HUANG, G. et al. High-efficient synthesis and biological activities of allosamidins. Journal of enzyme inhibition and medicinal chemistry, v. 30, n. 6, p. 863-866, 2015. HUANG, G.; HUANG, H.. Synthesis, antiasthmatic, and insecticidal/antifungal activities of allosamidins. Journal of enzyme inhibition and medicinal chemistry, v. 34, n. 1, p. 1226- 1232, 2019. HUANG, G.; SHU, S.; CHENG, F. Synthesis of Allosamidin Analogue Di-N-acetyl-β- chitobiosyl N-Glycoaminooxazoline. Letters in Drug Design & Discovery, v. 10, n. 8, p. 738-740, 2013. HULL, C. M.; DAVIDSON, R. C.; HEITMAN, J. Cell identity and sexual development in Cryptococcus neoformans are controlled by the mating-type-specific homeodomain protein Sxi1α. Genes & development, v. 16, n. 23, p. 3046-3060, 2002. GÓES, J. A. Preparações semissólidas contend extrato das flores de Allamanda cathartica L. com potencial antioxidante e anti-inflamatório, Universidade do Vale do Itajaí, Itajaí, 2011. JARVIS, J. N. et al. Ambition-cm: high-dose liposomal amphotericin for HIV-related cryptococcal meningitis. In: Proceedings of the Conference on Retroviruses and Opportunistic Infections, Seattle, WA, USA. 2017. p. 13-16. JARVIS, J. N., MEINTJES, G., WILLIAMS, A., BROWN, Y., CREDE, T., & HARRISON, T. S. Adult meningitis in a setting of high HIV and TB prevalence: findings from 4961 suspected cases. BMC infectious diseases, 10(1), 67, 2010. KALYANARAMAN, Chakrapani; BERNACKI, Katarzyna; JACOBSON, Matthew P. Virtual screening against highly charged active sites: identifying substrates of alpha− beta barrel enzymes. Biochemistry, v. 44, n. 6, p. 2059-2071, 2005. KARLSSON, M.; STENLID, J. Comparative evolutionary histories of the fungal chitinase gene family reveal non-random size expansions and contractions due to adaptive natural selection. Evolutionary Bioinformatics, v. 4, p. EBO. S604, 2008. KORB, O., MONECKE, P., HESSLER, G., STUTZLE, T.; EXNER, T. E. pharmACOphore: multiple flexible ligand alignment based on ant colony optimization. Journal of chemical information and modeling, 50(9), 1669-1681, 2010. KUBINYI, Hugo. Chemical similarity and biological activities. Journal of the Brazilian Chemical Society, v. 13, n. 6, p. 717-726, 2002. KURANDA, M. J.; ROBBINS, Phillips W. Chitinase is required for cell separation during growth of Saccharomyces cerevisiae. Journal of Biological Chemistry, v. 266, n. 29, p. 19758-19767, 1991. KURTZ, M. B., & DOUGLAS, C. M. Lipopeptide inhibitors of fungal glucan synthase. Journal of Medical and Veterinary Mycology, 35(2), 79-86. 1997. KWON-CHUNG, Kyung J. et al. Cryptococcus neoformans and Cryptococcus gattii, the etiologic agents of cryptococcosis. Cold Spring Harbor perspectives in medicine, v. 4, n. 7, p. a019760, 2014. LAM, CHI‐LEUNG et al. Pulmonary cryptococcosis: A case report and review of the Asian‐ Pacific experience. Respirology, v. 6, n. 4, p. 351-355, 2001. LANIADO-LABORÍN, R.; CABRALES-VARGAS, M. N. Amphotericin B: side effects and toxicity. Revista iberoamericana de micología, v. 26, n. 4, p. 223-227, 2009. LATGÉ, J. P. The cell wall: a carbohydrate armour for the fungal cell. Molecular microbiology, 66(2), 279-290. 2007. LAVECCHIA, Antonio; DI GIOVANNI, Carmen. Virtual screening strategies in drug discovery: a critical review. Current medicinal chemistry, v. 20, n. 23, p. 2839-2860, 2013. LEENDERS, A. C., REISS, P., PORTEGIES, P., CLEZY, K., HOP, W. C., HOY, J., ... & KROON, F. P. Liposomal amphotericin B (AmBisome) compared with amphotericin B both

29 followed by oral fluconazole in the treatment of AIDS-associated cryptococcal meningitis. Aids, 11(12), 1463-1471, 1997. LIN, X. Cryptococcus neoformans: morphogenesis, infection, and evolution. Infection, Genetics and Evolution, v. 9, n. 4, p. 401-416, 2009. LIN, Ying-Ying; SHIAU, Stephanie; FANG, Chi-Tai. Risk factors for invasive Cryptococcus neoformans diseases: a case-control study. PLoS One, v. 10, n. 3, p. e0119090, 2015. LIU, Tong-Bao; PERLIN, David S.; XUE, Chaoyang. Molecular mechanisms of cryptococcal meningitis. Virulence, v. 3, n. 2, p. 173-181, 2012. MADA, P. K., JAMIL, T., & ALAM, M. U. Cryptococcus (Cryptococcosis). In StatPearls [Internet]. StatPearls Publishing. 2019. MALIGIE, M. A., SELITRENNIKOFF, C. P. Cryptococcus neoformans resistance to echinocandins: (1,3)beta-glucan synthase activity is sensitive to echinocandins. Antimicrob Agents Chemother. 49:2851–2856, 2005. MCCREATH, K. J.; SPECHT, C. A.; ROBBINS, P. W. Molecular cloning and characterization of chitinase genes from Candida albicans. Proceedings of the National Academy of Sciences, v. 92, n. 7, p. 2544-2548, 1995. MEYER, W., AANENSEN, D. M., BOEKHOUT, T., COGLIATI, M., DIAZ, M. R., ESPOSTO, M. C., ... & LITVINTSEVA, A. P. Consensus multi-locus sequence typing scheme for Cryptococcus neoformans and Cryptococcus gattii. Medical mycology, 47(6), 561-570, 2009. MOODLEY, A. et al. Early clinical and subclinical visual evoked potential and Humphrey's visual field defects in cryptococcal meningitis. PLoS One, v. 7, n. 12, p. e52895, 2012. MOURAD, A.; PERFECT, J. R. The war on cryptococcosis: A Review of the antifungal arsenal. Memórias do Instituto Oswaldo Cruz, v. 113, n. 7, 2018. MUZZARELLI, R. A. A. Chitin. Pergamon Press Ltd. Headington Hill Hall, Oxford, England Pp.5-37, 1977. NACZK, M.; SHAHIDI, F. Phenolics in cereals, fruits and vegetables: Occurrence, extraction and analysis. Journal of pharmaceutical and biomedical analysis, v. 41, n. 5, p. 1523-1542, 2006. OMONHINMIN, A. C.; IJEOMA, P. D.; UCHE, A. In vivo antioxidant assessment of two antimalarial plants Allamamda cathartica and Bixaorellana. Asian Pac J Trop Biomed 2013; 3(5): 388-94 ONISHI, J., MEINZ, M., THOMPSON, J., CUROTTO, J., DREIKORN, S., ROSENCACH, M., ... & CABELLO, A. Discovery of novel antifungal (1, 3)-β-D-glucan synthase inhibitors. Antimicrobial Agents and Chemotherapy, 44(2), 368-377, 2000. ORNER, E. P. et al. Cell Wall-Associated Virulence Factors Contribute to Increased Resilience of Old Cryptococcus neoformans Cells. Frontiers in Microbiology, v. 10, p. 2513, 2019. OSTROSKY-ZEICHNER, L. et al. An insight into the antifungal pipeline: selected new molecules and beyond. Nature reviews Drug discovery, v. 9, n. 9, p. 719, 2010. OYELEYE, A.; NORMI, Y. M. Chitinase: diversity, limitations, and trends in engineering for suitable applications. Bioscience reports, v. 38, n. 4, 2018. PAPPAS, P. G. et al. A phase II randomized trial of amphotericin B alone or combined with fluconazole in the treatment of HIV-associated cryptococcal meningitis. Clinical Infectious Diseases, v. 48, n. 12, p. 1775-1783, 2009. PERFECT, J. R. et al. Clinical practice guidelines for the management of cryptococcal disease: 2010 update by the Infectious Diseases Society of America. Clinical infectious diseases, v. 50, n. 3, p. 291-322, 2010. PETRICEVICH, V. L.; ABARCA-VARGAS, R. Allamanda cathartica: A Review of the Phytochemistry, Pharmacology, Toxicology, and Biotechnology. Molecules 2019, 24, 1238.

PIERINI, L. M.; DOERING, T. L. Spatial and temporal sequence of capsule construction in Cryptococcus neoformans. Molecular microbiology, v. 41, n. 1, p. 105-115, 2001. PINZI, L.; RASTELLI, G. Molecular Docking: Shifting Paradigms in Drug Discovery. International journal of molecular sciences, v. 20, n. 18, p. 4331, 2019. POLACHECK, I.; HEARING, V. J.; KWON-CHUNG, K. J. Biochemical studies of phenoloxidase and utilization of catecholamines in Cryptococcus neoformans. Journal of bacteriology, v. 150, n. 3, p. 1212-1220, 1982. PYRGOS, V. et al. Epidemiology of cryptococcal meningitis in the US: 1997–2009. PLoS ONE 8, e56269 (2013). RAJASINGHAM, R. et al. Global burden of disease of HIV-associated cryptococcal meningitis: an updated analysis. The Lancet infectious diseases, v. 17, n. 8, p. 873-881, 2017. RANDHAWA, HARBANS S.; PAL, MAHENDRA. Occurrence and significance of Cryptococcus neoformans in the respiratory tract of patients with bronchopulmonary disorders. Journal of clinical microbiology, v. 5, n. 1, p. 5-8, 1977. RODRIGUES, M. L., ALVAREZ, M., FONSECA, F. L., & CASADEVALL, A. Binding of the wheat germ lectin to Cryptococcus neoformans suggests an association of chitinlike structures with yeast budding and capsular glucuronoxylomannan. Eukaryotic cell, 7(4), 602- 609. 2008. ROEMER, T., & KRYSAN, D. J. Antifungal drug development: challenges, unmet clinical needs, and new approaches. Cold Spring Harbor perspectives in medicine, 4(5), a019703, 2014. RONCERO, C. The genetic complexity of chitin synthesis in fungi. Current genetics, 41(6), 367-378, 2002. RUSH, C. L. et al. Natural product–guided discovery of a fungal chitinase inhibitor. Chemistry & biology, v. 17, n. 12, p. 1275-1281, 2010. SAGUEZ, J.; VINCENT, C.; GIORDANENGO, P. Chitinase inhibitors and chitin mimetics for crop protection. Pest technology, v. 2, n. 2, p. 81-86, 2008. SAKANE, M.; SHEPHERD, G. J. Uma revisão do gênero Allamanda L. (Apocynaceae). Revista Brasileira de Botânica, v. 9, n. 2. pg. 125-149, 1986. SAKUDA, S.; ISOGAI, A.; MATSUMOTO, S.; SUZUKI, A. Search for microbial insect growth regulators. II. Allosamidin, a novel insect chitinase inhibitor. Journal of Antibiotics 40, 296-300, 1987. SAKUDA, S.; SAKURADA, M. Preparation of biotinylated allosamidins with strong chitinase inhibitory activities. Bioorganic and Medicinal Chemistry Letters 8, 2987-2990, 1998. SALEEM, Muhammad et al. Isolation and characterization of secondary metabolites from Plumeria obtusa. Journal of Asian natural products research, v. 13, n. 12, p. 1122-1127, 2011. SANTOS-GANDELMAN, J.; MACHADO-SILVA, A. Drug development for cryptococcosis treatment: what can patents tell us?. Memórias do Instituto Oswaldo Cruz, 114, 2019. SEIDL, V. Chitinases of filamentous fungi: a large group of diverse proteins with multiple physiological functions. Fungal Biology Reviews, 22(1), 36-42, 2008. SELVAGGINI, S. et al. Independent regulation of chitin synthase and chitinase activity in Candida albicans and Saccharomyces cerevisiae. Microbiology, v. 150, n. 4, p. 921-928, 2004. SHEEHAN, D. J.; HITCHCOCK, C. A.; SIBLEY, C. M. Current and emerging azole antifungal agents. Clinical microbiology reviews, v. 12, n. 1, p. 40-79, 1999. SONG, C. M.; LIM, S. J.; TONG, J. C. Recent advances in computer-aided drug design. Briefings in bioinformatics, v. 10, n. 5, p. 579-591, 2009.

TIWARI, T. N.; PANDEY, V. B.; DUBEY, N. K. Plumieride from Allamanda cathartica as an antidermatophytic agent. Phytotherapy Research, v. 16, n. 4, p. 393-394, 2002. VASSEUR, R.; BAUD, S.; VIGOUROUX, X. A Framework for Inverse Virtual Screening Large-Scale Protein Targets Identification. BIOTECHNO 2014: The Sixth International Conference on Bioinformatics, Biocomputational Systems and Biotechnologies. ISBN: 978-1- 61208-335-3. 2014 VELAGAPUDI, R. et al. Spores as infectious propagules of Cryptococcus neoformans. Infection and immunity, v. 77, n. 10, p. 4345-4355, 2009. VILJOEN, A.; MNCWANGI, N.; VERMAAK, I. Anti-inflammatory iridoids of botanical origin. Current medicinal chemistry, v. 19, n. 14, p. 2104-2127, 2012. WANG, J. et al. Primary cutaneous cryptococcosis treated with debridement and fluconazole monotherapy in an immunosuppressed patient: a case report and review of the literature. Case reports in infectious diseases, v. 2015, 2015. WHO - World Health Organization. Rapid Advice: diagnosis, prevention and management of cryptococcal disease in HIV-infected adults, adolescents and children (2011). Geneva: World Health Organization. WILLIAMSON, P. R., JARVIS, J. N., PANACKAL, A. A., FISHER, M. C., MOLLOY, S. F., LOYSE, A., & HARRISON, T. S. Cryptococcal meningitis: epidemiology, immunology, diagnosis and therapy. Nature Reviews Neurology, 13(1), 13, 2017. WILLIAMSON, P. R. Biochemical and molecular characterization of the diphenol oxidase of Cryptococcus neoformans: identification as a laccase. Journal of bacteriology, v. 176, n. 3, p. 656-664, 1994. WIRTH, F.; AZEVEDO, M. I.; GOLDANI, L. Z. Molecular types of Cryptococcus species isolated from patients with cryptococcal meningitis in a Brazilian tertiary care hospital. Brazilian Journal of Infectious Diseases, v. 22, n. 6, p. 495-498, 2018. XU, X.; HUANG, M.; ZOU, X. Docking-based inverse virtual screening: methods, applications, and challenges. Biophysics reports, v. 4, n. 1, p. 1-16, 2018. ZARAGOZA, O. et al. The capsule of the fungal pathogen Cryptococcus neoformans. Advances in applied microbiology, v. 68, p. 133-216, 2009. ZARAGOZA, O. Basic principles of the virulence of Cryptococcus. Virulence, v. 10, n. 1, p. 490-501, 2019. ZHU, Z. et al. Acidic mammalian chitinase in asthmatic Th2 inflammation and IL-13 pathway activation. Science, v. 304, n. 5677, p. 1678-1682, 2004.

CAPÍTULO I

Manuscrito: A plumieridine-rich fraction from Allamanda polyantha inhibits chitinolytic activity and exhibits antifungal properties against Cryptococcus neoformans Artigo submetido ao periódico Frontiers in Microbiology

A plumieridine-rich fraction from Allamanda polyantha inhibits chitinolytic activity and exhibits antifungal properties against Cryptococcus neoformans

Eden Silva e Souza1, Vanessa de Abreu Barcellos2, Júlia Catarina Vieira Reuswaat2, Nicolau Sbaraini2, Rafael de Oliveira Schneider2, Ane Wichine Acosta Garcia 2, Gilsane Lino von Poser3, Euzébio Guimarães Barbosa4, João Paulo Matos Santos Lima1, Marilene Henning Vainstein2

1Bioinformatics Multidisciplinary Environment – BioME, Universidade Federal do Rio Grande do Norte, Natal, Brazil 2Centro de Biotecnologia, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil 3Department of Pharmacy, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil 4Department of Pharmacy, Universidade Federal do Rio Grande do Norte, Natal, Brazil

* Correspondence: Marilene Henning Vainstein

Keywords: Cryptococcus neoformans; target prediction; chitinase; plumieridine; antifungal activity; drug discovery; glycoside hydrolase family 18.

Abstract Cryptococcosis is a fungal infection caused by some Cryptococcus species (Cryptococcus gattii and Cryptococcus neoformans are the main causative agents). The infection initiates when propagules are inhaled, with consequent deposition into the lung alveoli. If not properly treated, Cryptococcus infection may reach the central nervous system and result in death. Cryptococcosis recommended treatment takes a 3-stage regimen and three drugs are employed: fluconazole, amphotericin B and flucytosine. Although effective, these drugs are unavailable worldwide, can lead to resistance development, and may display toxic effects to the patients. Thus, new drugs for cryptococcosis treatment are needed. A recently found iridoid for the Allamanda genus, named plumieridine, derives from Allamanda polyantha seed extracts and has shown remarkable antifungal activity against Cryptoccoccus neoformans with a MIC of 0.250 mg/mL. In order to address the mode of action of plumieridine several in silico and in vitro experiments were performed. Using a virtual screening approach, a chitinase was identified as a potential target. Noteworthy, C. neoformans cell-free supernatant incubated with plumieridine displayed reduced chitinase activity, while chitinolytic activity was not inhibited in the insoluble cell-related fraction. Additionally, confocal microscopy revealed changes in the distribution pattern of chitooligomers in the cell wall, from a polarized to a diffuse cell pattern state. Remarkably, further assays shown that plumieridine can also inhibit the chitinolytic activity from supernatant and cell-free extracts of bacteria, insect and mouse cells. The results confirm the virtual screening target prediction and suggest that plumieridine can be broad spectrum inhibitor for the glycoside hydrolase 18 superfamily.

1. Introduction Cryptococcosis is a neglected fungal infection caused by species of Cryptococcus genus, mostly by strains in the complex Cryptococcus gattii-Cryptococcus neoformans (Kwon-Chung, et al., 2017). The infection occurs when cryptococci dried cells or spores are inhaled and reach the lungs, where it may remain latent for a certain period (Velagapudi et al., 2009; Giles et al., 2009). If the infection is not properly treated, Cryptococcus cells disseminates through the

34 central nervous system, where it may lead to fatal meningitis (Srikanta et al., 2014). Cases of cryptococcosis are closely related to the pandemic cases of AIDS, with less than 300 reports in the 1950s and more than a million in 2006 (Park et al., 2009). However, the Joint United Nations Programme on HIV/AIDS (UNAIDS) estimates that 1.4 million AIDS patients still die annually and 15% of these deaths are caused by cryptococcosis (Rajasingham et al., 2017). Moreover, reports of cryptococcosis in immunocompetent patients are increasing (Poley et al., 2019; Chen et al., 2008; Suchitha et al., 2012). Treatment of cryptococcal meningitis consists of an induction, consolidation, and maintenance regimen (Mourad; Perfect, 2018b). The Infectious Disease Society of America (IDSA) recommends a two-week treatment with amphotericin B (AMB) and flucytosine (5-FC) followed by treatment with fluconazole with time and dose depending of the patient’s response (Perfect et al., 2010). This combination of AMB and 5-FC shows more fungicidal activity than the sole treatment with AMB, as it was administered earlier (Sloan; Parris, 2014). Although not always effective, this is still the best treatment available nowadays but, unfortunately, it is not commercialized worldwide (Pappas, 2010). Additionally, the drugs currently used in cryptococcosis treatment have some disadvantages, for instance, the administration of fluconazole as monotherapy or at the end of AMB-5-FC treatment, may lead to fungal heteroresistance (Santos-Gandelman; Machado-Silva, 2019; Coelho; Casadevall, 2016; Stone et al., 2019). Furthermore, hepato- and nephrotoxicity are also reported during the cryptococcosis treatment, a side effect mostly caused by AMB (Krysan, 2015). In this context, natural products are an interesting starting point. The basis of most medicines comes from studies employing natural sources (Atanasov et al., 2015) and drugs with less/no toxicity, but still effective against Cryptococcus sp. are needed. Consequently, studies of natural compounds with anti-cryptoccocal activity are increasing (da Silva et al., 2016; Teixeira et al., 2018). Cryptococcus cells have two relevant therapy targets: the extracellular polysaccharide capsule and the cell wall. The cryptococci capsule is composed of the polysaccharides glucuronoxylomannan (GXM) and galactoxylomannan (GalXM) and mannoproteins. GXM specific antibodies have been observed to cause interference in the building of capsular fibrils, thus it may contribute to host defence (Martinez, et al., 2004; Agustinho et al., 2018). The cell wall is composed of β-linked glucans [β-(1,3) and β-(1,6)] and chitin (i. e., accounts for the rigidity and integrity of the cell wall), a linear polymer of β-1,4-N-acetyl-glucosamine (GlcNAc) (Gilbert et al., 2010, Agustinho; Nosanchuk 2017; Banks et al. 2005). Echinocandins are important antifungal agents capable of interrupting the β-(1,3) glucan synthesis, being an effective treatment against several fungal infections (Denning, 2003). However, none of the discovered echinocandins is effective against C. neoformans (Feldmesser et al., 2000; Maligie; Selitrennikoff, 2005). Chitinases are enzymes responsible for the hydrolysis of the β-1,4 linkage in chitin resulting in monomers and oligomers of GlcNAc (Howard et al., 2003). These enzymes have been reported in a variety of organisms, which includes fungi (Junges et al., 2014), plants (Grover, 2012), bacteria and humans (Rathore; Gupta, 2015). Chitinases can be classified into endochitinases and exo-chitinases. The first breaks chitin randomly at internal sites and releases low molecular mass multimers of GlcNAc, for example chitotriose (Sahai; Manocha, 1993). The latter is divided into chitobiosidases (β-N-acetylhexosaminidase), which catalyzes di- acetylchitobiose starting at the non-reducing end of chitin, and 1-4-β-glucosaminidase, that is responsible to cleave oligomeric products of endochitinases and, thus generates GlcNAc monomers (Liu; Kokare, 2017). Regarding the classification, the Carbohydrate-Active enZYmes Database (CAZy) groups chitin degrading enzymes into the Glycoside Hydrolase families (GH) 18, 19 and 20 (Lombard et al., 2014). GH18 and 19 are known as chitinases due to the ability of degrading chitin polymers, while GH20 cleaves dimeric units of GlcNAc and comprises the ezymes chitobiases

35 and β-N-acetylhexosaminidases (Funkhouser; Aronson, 2007; Oyeleye; Normi, 2018). All fungi are reported to have chitinases of the glycoside hydrolase 18 family (GH18) (Oyeleye; Normi, 2018), excepting the parasitic fungus Nosema bombycis, which harbors GH19 chitinases (Han et al., 2016). Controversially, only chitinases from family GH19 are observed in plants (Prakash et al., 2010). In fungi, chitinases are known to participate on hyphal branching and growth, autolysis, and morphogenesis (Duo-Chuan, 2006). In C. neoformans, chitinases have been reported to be essential for virulence and during sexual reproduction, however, they are not required for asexual reproduction (Baker et al., 2009). Several chitinase inhibitors have been described in the literature. For instance, the natural peptides argifin and argadin can inhibit Aspergillus fumigatus, Serratia marcescens, and human GH18 chitinases (Rao et al., 2005a). Caffeine was shown to be a chitinase inhibitor for the fungus Clonostachys rosea chitinase (CrChi1). Also, conservation in the binding site is crucial for the effectiveness of this inhibitor (Yang et al., 2010). Methylxanthine, which includes caffeine, is a drug class used as inflammatory agents, and are also reported as GH18 chitinase inhibitors (Rao et al., 2005b). A chemical class named acetazolamide has been reported as a chitinase inhibitor in the pathogenic fungus A. fumigatus (Schüttelkopf et al., 2010). Noteworthy, several chitinolytic enzymes are not intracellular, what makes it possible to explore inhibitors that do not need to cross the cell wall and plasmatic membrane (Hamid et al., 2013). The Allamanda genus (Apocynaceae: Gentianales) comprises 15 plant species distributed in South America (Sakane; Shepherd, 1986; The Plant List, 2019). Included in this genus, Allamanda polyantha is endemic of the Brazilian Mata Atlântica rain forest (Flora do Brasil, 2019). This plant genus is used in the popular medicine to treat several illnesses, with potential antifungal, diuretic, antidiabetes and antiparasitic properties (Petricevich; Abarca-Vargas, 2019). Several iridoid compounds have been isolated from Allamanda spp., especially from A. cathartica. The iridoids found in A. cathartica include, but are not limited to, plumiericin, isoplumiericin, plumieride, and plumieride coumarate (Petricevich; Abarca-Vargas, 2019). However, the aglycone configuration of plumieride, plumieridine, has not been reported for the Allamanda spp. On the other hand, isolation of plumieridine was reported from Plumeria obtusa, another Apocynaceae (Saleem et al., 2011). Recently, our research group identified the anticryptococcal activity in the seed’s extract of A. polyantha Müll. Arg. and this activity was attributed to a fraction rich in the monoterpenoid plumieridine. However, plumieridine’s target and mode of action are unknown. We hypothesized that the natural compound plumieridine is the antifungal agent. In order to test this hypothesis a ligand based virtual screening was performed and indicated that plumieridine targets Cryptococcus neoformans chitinases. Thus, in this study, several in vitro and in silico assays were employed to evaluate the mechanism of action of this molecule and possible interaction with C. neoformans chitinases. Furthermore, the activity of plumieridine against mouse, insect and bacteria chitinases was also evaluated.

2. Material and methods 2.1 Plumieridine isolation and Nuclear magnetic resonance Seeds crushed in a blender were placed in contact with ultrapure water (10 g/20 mL) for 4 h, under agitation. The liquid suspension was centrifuged (for 10 min at 7168 x g). The resulting supernatant was filtered in filter paper and polypropylene prefilter (AP 25, Millipore). The aqueous extract was completely lyophilized at -50 °C and 0.040 mbar (Christ Alpha 1-4 LD plus, Germany) and stored at -80 °C. Lyophilized crude extract was subjected to silica gel column chromatography (70-320 mesh, Merck), using a gradient elution of dichloromethane: methanol (95:5 to 80:20) as the mobile phase to obtain plumieridine. The fractions were chromatographed over preparative TLC (20 cm × 20 cm, 0.5 mm layer, SiO2 F254 plates –

Merck) using a mixture of dichloromethane:methanol (80:20) as eluent. Fractions were subjected to another chromatographic column as above mentioned in order to obtain the most purified compound. Fractions with antifungal activity were submitted to nuclear magnetic resonance (NMR) recorded in CD3OD, on a Varian spectrometer, operating at 400 MHz. Peaks of residual water were used as an internal standard in 1H NMR spectra and solvent peak was used as internal standard in 13C spectra (Gottlieb et al., 1997). Results were analyzed with MestraNova software (v. 6.0.2). 2.2 Antifungal susceptibility assay Minimum inhibitory concentration (MIC) of the plumieridine was determined against C. neoformans according to the Clinical and Laboratory Standards Institute M27-A2. The compound was resuspended in Milli-Q water with 10% DMSO, and filtered prior to use (polyvinylidene difluoride filter, 0.22 μm pore size, Millipore). MIC assay was performed in 96-well plate, where plumieridine was serially diluted, starting with 20 mg/mL, in RPMI 1640 (pH 7; 2% glucose) buffered with MOPS. Plates were incubated at 37 °C for 72 h. 2.3 Virtual screening In order to predict plumieridine’s targets an ad hoc ligand based virtual screening approach was performed. PharmACOphore, which allows the alignment of active compounds, was used to search for similar ligands bound to proteins in the Protein Data Bank (PDB) (Hessler et al., 2010). 2.4 Molecular modeling and docking Virtual screening pointed a chitinase from A. fumigatus (PDB 3CHE) as a putative target, and, to date, the structures from C. neoformans chitinases are not resolved experimentally. Firstly, chitinase sequences from C. neoformans, previously identified by Baker and collaborators (2009), were retrieved from FungiDB (Basenko et al., 2018) under the access codes: CNAG_03412 (CHI2), CNAG_02598 (CHI21), CNAG_04245 (CHI22), and CNAG_02351 (CHI4). Evaluation on sequences’ signal peptides, transmembrane helices, and conserved domains were performed with SignalP, TMHMM, and Conserved Domain Database (CDD), respectively (Armenteros et al., 2019; Möller et al., 2002; Li et al., 2020). As the best- identity hit for C. neoformans chitinase sequences against potential PDB templates were around 30 %, molecular models were created using different approaches. Sequences were modeled on SwissModel (Waterhouse et al., 2018), Phyre2 (Kelley et al., 2015) and Robetta server (Song et al., 2013). All models were evaluated on SwissModel Structure Assessment Tool and the best model for each chitinase chosen based on Ramachandran-favored, Outliers, MolProbity Score, QMEAN, and Rotamer Outliers (Benkert et al., 2011). Chitinase models were lately used for the molecular docking and dynamics simulations. Molecular docking of all four chitinases and plumieridine was simulated using AutoDock Vina (Trott; Olson, 2010), with UCSF Chimera interface (Pettersen et al., 2004). The position of the ligand identified in virtual screening was considered as the lowest energy pose, thus, plumieridine was manually docked in the same position in all C. neoformans chitinases. In order to infer the way plumieridine interacts in the active site, two orientations were assayed arbitrarily named inward and outward (Supplementary Material Figure S2) 2.5 Molecular dynamics simulations Molecular dynamics simulations were performed on complexes obtained from molecular docking using GROMACS 5 software (Van Der Spoel et al., 2005) with the aid of CHARMM force field (Vanommeslaeghe et al., 2010). Plumieridine-chitinase complexes were placed inside a cubic box large enough to let minimum 1.0 nm space from the protein to the box and the solvent properties were mimetic using TIP3P water model. The system had the charge neutralized with the addiction of ions at physiological concentration (0.15 µM). Volume (NVT) and pressure (NPT) equilibrium simulations were geometrically optimized in the solvated system. During the simulation, the temperature was constant at 300 K coupling the system with

V-rescale thermostat with a 13-coupling time of 0.1 ps. The pressure was also kept constant at 1 bar with Parinello-Rahman coupling algorithm. Molecular dynamics simulations were performed during 2600 ps. ensuring stabilization of root-mean-square deviation (RMSD). 2.6 Chitinase activity and inhibitory assays In order to evaluate plumieridine’s inhibitory activity on chitinase we used a variety of models and fractions. These models included C. neoformans, Bacillus subtilis, Tenebrio molitor, and macrophages lineage J774.A1. C. neoformans was grown on either YPD (1% yeast extract, 2% dextrose, and 2% Bacto peptone) or YPGlcNAc (1% yeast extract, 2% N- acetylglucosamine, and 2% Bacto peptone) on shaker for 24 h at 30 °C. Both media were used to compare whether there is a difference in chitinase activity due to change in the carbon source (Baker et al., 2009). After incubation, the culture was centrifuged (9000 x g for 10 min) and the culture supernatant was collected and lyophilized (secreted fraction). Lysis buffer (50 mM Tris- HCl pH 8.0, 20 mM EDTA, 200 mM NaCl, 1% SDS, and 1% triton) was used to release the proteins attached to plasmatic membrane. After centrifugation, two fractions were collected and named soluble cell fraction and insoluble cell fraction. All samples were lyophilized and solubilized in PBS buffer (20 mg/mL). B. subtilis was grown on LB medium at 37 °C for 24 h. The supernatant was lyophilized and resuspended to the concentration of 2 mg/mL in PBS. This solution was used in the assay. In order to evaluate the plumieridine’s activity against chitinases from insects, the mealworm larvae (T. molitor) was employed. Eight and half grams of whole T. molitor larvae were dried freeze and ground to powder using liquid nitrogen. The powder was homogenized in PBS 1:2 (W/V) for 15 min under agitation. Tenebrio’s crude extract was centrifuged (9000 x g for 10 min) and filtered, the resulting supernatant was lyophilized and, then, resuspended in PBS (2 mg/mL). Lastly, J774.A1 cells, obtained from Banco de Células do Rio de Janeiro (BCRJ; accession number 0121) were cultured in DMEM (Dulbecco’s modified Eagle’s medium; Gibco® Life Technologies, MA, USA) supplemented with 10% heat-inactivated fetal bovine serum (FBS; Gibco Life Technologies), 1 mM L-glutamine (Gibco Life Technologies), 1 mM sodium pyruvate (Gibco Life Technologies), 1% nonessential amino acids (Gibco Life Technologies) and incubated at 37 °C and 5% CO2 for three days. After this step, the cell culture was centrifuged (9000 x g for 10 min) and the resulting supernatant was lyophilized and, then, resuspended in PBS (20 mg/mL). Chitinase activity assays were performed employing 4-metilumbeliferil-β-D-NNN-triacetil chitotriose (Sigma) as substrate. A standard curve was created using 4-methylumbelliferyl (4MU). The assay was performed in a 96-well coated microplate and consisted of 100µL of McIlvaine buffer pH 6,0, 5 µL of substrate, and 10µL of sample. The reaction was incubated at 37°C and the fluorescence was read at 355 nm excitation and 460 nm emission on SpectraMax I3. Inhibitory assays employed an increasing plumieridine concentration diluted in McIlvaine buffer pH 6,0. Plumieridine was added in the following concentrations: 0, 33, 100, 160, and 260 µg/mL in a final volume of 200 µL. Quantification of samples were performed according to relative fluorescent units (RFU), using the standard curve previously created. 2.7 Real time quantitative PCR (RT-qPCR) Fungi were grown in the same media previously described for 4 h at 30°C and 200 rpm in the presence and absence of plumieridine using the sublethal dose determined through MIC. Total RNA extraction was performed using glass beads and Trizol (Invitrogen) treatment following manufacturer's recommendations. cDNA was synthesized using 1 µg of DNase- treated RNA and ImProm-II™ Reverse Transcription System (Promega) following manufacturer's recommendations. PCR reactions were conducted at a final volume of 20 µL, containing 2 µL of the cDNA (4 ng/µL), 2 µL SYBR Green (1:1000) (Invitrogen), 0.1 µL dNTP (5mM), 2µL PCR buffer 10x, 1,2µL MgCl2, 0.05 U Platinum Taq DNA Polymerase (Invitrogen), and 0,2 µL of each primer (5 pmol).The experiment was carried out on an Applied Biosystems 7500 Fast Real-Time PCR System® with thermal cycling conditions set with an

38 initial step at 95°C for 10 min, followed by 40 cycles at 95°C for 15 s, 60°C for 15 s and 72°C for 60 s. A melting curve analysis was performed at the end of the reaction to confirm the presence of a single PCR product. All experiments were performed using three independent cultures, and each cDNA sample was analyzed in triplicate for each primer pair. Quantifications were normalized to an actin cDNA expression used in the experiments. C. neoformans’ GH18 chitinase primers used were described by Baker et al. (2009) (STable S1). 2.8 Confocal microscopy Yeast cells were grown on YPD and in the absence and presence of plumieridine. Cells were treated with the following concentration of the compound: 0 µg/mL, 150 µg/mL, 300 µg/mL, and 600 µg/mL and incubated in minimal medium for 24 h at 37°C with 5% CO2. Cells were washed with PBS and fixed with paraformaldehyde, followed by incubation with 5 µg/mL of Wheat Germ Agglutinin (WGA) conjugated with Alexa 488 (green) for 30 min at 37 °C. Cells were washed again with PBS and incubated with Calcofluor white (blue) for 30 min at 37 °C. As treatment with plumieridine reduces cell count, it was not possible to observe the impact of treatment with different drug concentrations in the same number of cells. Cell count was performed and a percentage ratio between the total cell count and cells observed with differences was calculated for all treatments. 2.9 Statistical analysis All experiments were performed in triplicate. One-way analysis of variance (One-way ANOVA) was used to evaluate triplicates from the same experiment, while Two-way ANOVA was used to perform comparisons between experiments. All graphs were generated in Prism – GraphPad 8.0 (GraphPad Software, Inc., San Diego, CA). Letters in the graphs indicate statistical significance between samples tested. 3. Results 3.1 Plumieridine targets chitinases Virtual screening resulted in 14 993 predicted targets, among these, 38 hits belonged to Aspergillus fumigatus (Table 1 and Supplementary Material Table S2). Results were ranked according to free energy and A. fumigatus chitinase B1 in complex with the tripeptide VR0 (PDB 3CHE) was considered as the most probable target for this fungal species (Andersen et al., 2008). Full results for A. fumigatus predicted targets are available in supplementary material (Table S2.). 38 hits were found for A. fumigatus, however due to multiple hits for the same target, only 19 are unique. For instance, the PDB IDs 4LNB and 4LNG were found two times, 3CHE, 1W9U, and 4C1Y three times, 1W9V and 3CHF were found four times, 4D52 was found 6 times. Fourteen of the nineteen hits are from chitinase structures, thus this results points chitinase a promising target for plumieridine.

Table 1. Virtual screening results. Five of 38 hits for Aspergillus fumigatus based on ligand molecular fingerprint. The most probable target based on A. fumigatus bound ligands is the chitinase B complexed with the tripeptide obtained from the pentapeptide argifin.

Protein PDB ID Ligand Chitinase 3CHE Tripeptide (VR0) Chitinase 3CHD Dipeptide (WRG) Farnesyltransferase 4LNG Farnesyldiphosphate and tipifarnib Chitinase 2IUZ C2-dicaffeine Chitinase 2A3B Caffeine

3.2 Plumieridine interacts with chitinase catalytic residues Signal peptide prediction analysis indicates that CHI4, CHI21, and CHI22 can be secreted, while CHI2 does not have a predicted signal peptide sequence (Supplementary Material Table S1). TMHMM analysis shows that only CHI2 has a transmembrane domain (Supplementary Material Table S1) that expands from residues 21 to 43, a feature previously observed by Baker and collaborators (2009). Best models were selected based on parameters presented by SwissModel Evaluation tool and all models presented 80 % or more residues in Ramachandran favored areas. All models used were created by SwissModel. Notably, a folding conservation in the active site of all four chitinases can be observed (Supplementary Material Figure S2). Molecular binding analysis indicates that plumieridine putatively interacts with amino acid residues in the active binding site. NCBI Conserved Domains analysis shows C. neoformans CHI4 and CHI2 present a GH18_chitinase-like domain (cl10447), while CHI21 and CHI22 have a Glyco_hydro_18 domain (cl23725) (Supplementary Material Table S1). Both domains are part of the Glycoside Hydrolase 18 superfamily. This superfamily presents the characteristic motif DxDxE observed through sequence alignment (Supplementary Material Figure S3) (Vaaje-Kolstad et al., 2004). It was observed in the binding simulations that plumieridine putatively interacts closely to the DxDxE motif. As these residues are directly involved in catalysis, the predicted interaction may be responsible for loss in the catalytic activity.

Figure 1. Plumieridine initial docked position. Plumieridine was manually docked taking as reference the position of the bound ligand found through virtual screening analysis. The inward and outward orientations of plumieridine were tested for the four Cryptococcus neoformans chitinases. Plumieridine is colored in red. 3.3 Plumieridine inhibits chitinase activity in soluble fractions of Cryptococcus neoformans Through NMR spectra, it was only possible to identify plumieridine’s structure in the chromatographic fraction of the seed extract from A. polyantha with anti-cryptococcal activity (Supplementary Material S5 and S6). Minimal inhibitory concentration of the plumieridine-rich fraction against C. neoformans was 0.250 mg/mL. By sequence and structure comparison, chitinase 42 from Trichoderma harzianum was chosen as positive control for chitinolytic activity assays (Supplementary Material S7). A solution of 1 mg/mL of the Lysing Enzymes from Trichoderma harzianum (Sigma-Aldrich) was used in each experiment as control. For C. neoformans, chitinolytic activity was significantly higher in the insoluble cell fraction, for cells

40 grown either on YPD or YPGlcNAc, when compared to the other fractions evaluated (Figure 2A and 2B). For fungal grown on YPD, the chitinolytic activity in the secreted fraction did not show statistical difference when compared to the soluble cell fraction. While fungal grown on YPGlcNAc presented higher chitinolytic activity in the soluble cell fraction than in the secreted fraction (Figure 2A and 2B). Secreted and soluble cell fractions of C. neoformans grown in both media showed significant reduction in chitinase activity in the presence of plumieridine. Noteworthy, the secreted fraction shows a constant reduction in the chitinase activity between the values of 33 µg/mL and 160 µg/mL was observed for C. neoformans grown in YPD, whereas reduction was observed between 33 µg/mL and 100 µg/mL for cell grown on YPGlcNAc (Figure 2C and 2D). Similarly, a reduction in chitinolytic activity in the soluble cell fraction was observed in the assays employing the maximum concentration of 260 µg/mL of plumiriedine for C. neoformans grown on YPD and YPGlcNAc (Supplementary Material Figure S8A, and S8B). Conversely, chitinolytic activity in the insoluble cell fraction dia not present the dose- dependent inhibition pattern. For cells in this fraction of fungal grown on YPD, the maximum inhibition of chitinolytic activity was observed with 100 µg/mL of plumieridine. While the insoluble cell fraction of fungal grown on YPGlcNAc needed 300 µg/mL to present a reduction in chitinolytic activity. (Supplementary Material Figure S8C, and S8D)

Figure 2. Cryptococcus neoformans chitinolytic activity. (A) Chitinolytic activity of Trichoderma harzianum Lysing Enzymes and fractions of Cryptococcus neoformans grown on YPD; (B) and YPGlcNAc; (C) inhibition of chitinolytic activity by plumieridine in the secreted fraction of fungal grown on YPD; (D) and YPGlcNAc. Letters above bars indicate statistical significance between different concentrations.

3.4 Transcriptional levels of CHI22 are reduced in the presence of plumieridine RT-qPCR revels that the most expressed chitinase gene in C. neoformans, when cells are grown on either YPD and YPGlcNAc, is CHI4. Though CHI4 showed the highest relative expression levels, it was reported that mutants expressing only CHI4 or CHI21 does not present chitinolytic activity (Baker et al., 2009). CHI2, CHI21, and CHI4 transcriptional levels were not influenced by the carbon sources tested (glucose and N-acetylglucosamine) or treatment with plumieridine (Supplementary Material Figure S9). Notably, CHI22 was the only chitinase which transcriptional levels were influenced by the carbon source: the relative expression levels when the fungus was grown in YPGlcNAc was significantly higher when compared to YPD. Remarkably, CHI22 expression levels of fungal grown on YPGlcNAc were significantly reduced when cells were grown in the presence of plumieridine (Figure 3C and 3D).

Figure 3. Relative expression of levels of Cryptococcus neoformans’ chitinases. Relative expression levels of the four chitinases identified in C. neoformans genome when the fungus was grown in YPD (A) and YPGlcNAc (B); (C) CHI22 expression levels for cells grown in the presence and absence of plumieridine and on YPD; (D) and YPGlcNAc. Letters above bars indicate statistical difference between treatments. 3.5 Plumieridine changes chitin oligomers patterns in Cryptococcus neoformans cells Confocal microscopy revealed what was previously observed through MIC assay: plumieridine reduces cell count in a dose-dependent manner (Figure 4). Cell counts of 31, 29, 23, and 7 cells per field were observed in control, 150, 300, and 600 µL/mL of plumieridine treatment, respectively. Cells treated with 300 e 600 µL/mL of plumieridine have incomplete mother-daughter separation, evidenced by a group of three cells in line (Figure 4, white arrow). Changes in chitin from control and treated cells were not observed through calcofluor white staining (Figure 4). However, WGA appears in one or, more frequently, two dots per cell, what we named of polarized pattern. Nonetheless, 6.4% (2/31) of the cell count present WGA

42 staining diffuse the cell wall (Figure 4). The diffuse staining pattern increases with higher plumieridine concentrations: 10% (3/29) of cells in control show a diffuse pattern of WGA staining, while 34% (10/29) of cells presented it in the treatment with 150 µg/mL, 43% (10/23) of cells presented it when cells were treated with 300 µg/mL and 57% (4/7) of the cells presented it when cells were treated with 600 µg/mL of plumieridine.

Figure 4. Confocal microscopy of Cryptococcus neoformans treated with different plumieridine concentrations. Panel A shows cells marked with calcofluor white (blue) and WGA (green). Panel B shows cells marked only with WGA (green). 3.6 Plumieridine exerts inhibitory activity against chitinases from GH18 superfamily Local alignment shown that human chitinase (PDB 1HKI, Rao et al., 2003) and mouse chitinase (PDB 1VF8, Tsai et al., 2004) possess 48% of identity and 67% of similarity (data not shown). Structure comparison reveals that these chitinase structures present a RMSD of 0.692 Å. Based on this RMSD value, murine macrophage (J774.A1) chitinases was employed as a model for mammal chitinolytic activity inhibition assay, and the results can potentially be applied for humans (Supplementary Material S11). Chitinase inhibitory activity assays employing J774.A1 supernatant and plumieridine revealed a constant reduction in the chitinase activity between the values of 33 to 260 µg/mL, whereas higher plumieridine concentrations failed to reduce the chitinolytic activity even more (Figure 5A). Similar results were found employing B. subtilis supernatant (Figure 5B), where chitinolytic activity was inhibited up to 260 µg/mL, with no further enhancement with a higher amount of plumieridine. T. molitor supernatant showed significant inhibition of chitinase activity in treatment employing the maximum plumieridine concentration of 160 µg/mL with no further enhancement with a higher concentration of the compound (Figure 5C). These three model organisms used possess GH18 chitinase. The fact plumieridine is able to inhibit the activity of different organisms' members of GH18 and that simulations show interaction with catalytic residues, this result suggests that plumieridine is, potentially, an inhibitor of GH18 superfamily.

Figure 5. Inhibition of chitinolytic activity by plumieridine in different models. (A) Inhibition of chitinolytic activity in the supernatant of macrophages lineage J774.A1; (B) Inhibition of chitinolytic activity in the supernatant of B. subtilis; (C) Inhibition of chitinolytic activity in the filtered crude extract of the mealworm larvae T. molitor. Letters above bars indicate statistical difference between different treatments. 3.7 Plumieridine interacts differently with Cryptococcus neoformans chitinases All plumieridine-chitinase complexes reached system equilibrium evidenced by RMSD analysis (Supplementary Material S12). Molecular dynamics reveals that the inward plumierine orientation is the most energetically favorable binding configuration. For CHI2-, CHI21- and CHI22-plumieridine complexes in the initial outward orientation, it was observed that plumieridine rotates and changes to the inward orientation in the end of the simulation (Figure 6/Panel B). It suggests that plumieridine only in the inward orientation is able to inhibit chitinolytic activity. Controversially, CHI4-plumieridine complexes in both orientations had the inhibitor expelled from the active site. It suggests plumieridine is prone to inhibit GH18 superfamily chitinases, however interactions happen according to affinity for residues in the active site (Figure 6). Plumieridine inhibits GH18 chitinases selectively and in different levels, thus, this result also explains the residual chitinase activity observed in higher concentration of the inhibitory assays.

Figure 6. Chitinase-plumieridine complex interaction after dynamics simulation with initial inward/outward ligand’s orientation. Panels with inward and outward plumieridine

44 orientations regarding the active site of each Cryptococcus neoformans chitinase. Plumieridine is shown in red. 4. Discussion There is a constant need for new, cheaper, less toxic, and widely available drugs for cryptococcosis treatment (Mourad; Perfect, 2018a; Mourad; Perfect, 2018b; Coelho; Casadevall, 2016). In this way, South America’s biodiversity can be a rich source of new molecules (Valli et al., 2012). Although originally isolated from P. obtusa, identification of plumieridine in A. polyantha (a plant native of South America) seed extracts is not surprising since other iridoids have been found in members of the Apocynaceae family, however this is the first report of this iridoid for Allamanda spp. (Silva et al., 2007; Abe et al., 1984; Coppen, 1983; Nahry et al., 2016; Prabhadevi et al., 2012). Furthermore, the promising MIC (0.250 mg/mL), displayed by plumieridine against C. neoformans, led us to investigate the potential drug-target interactions. The time from the discovery and trials of a potential drug is estimated to be around 14 years (Song et al., 2009) and costs US$ 800 million (Lavecchia; Di Giovanni, 2013). Additionally, detailed information about drug-target interactions can also consume several years. Virtual screening approaches (as those applied here to identify chitinases as potential plumieridine’s targets) can reduce substantially the research time, providing detailed information on drug- target interactions (Kitchen et al., 2004). In order to prove the results found through the virtual screening approaches several chitinolytic assays were conducted. Remarkably, relative chitinolytic activity in C. neoformans-soluble fractions was reduced by the presence of plumieridine, proving the efficiency of our approach. Although previous reports have pointed that chitinases are not required to asexual reproduction in C. neoformans KN99a and KN99α (Baker et al., 2009), our results suggest that partial impairment of chitinolytic activity can lead to cell death through defects in asexual reproduction. When chitinase activity is absent, cell aggregation and incomplete cytokinesis can be observed, as it was noticed when cryptococcoci cells were treated with plumieridine. In spite of the fact that in silico approaches have predicted that plumieridine would be active against all C. neoformans chitinases, chitinolytic activity in the insoluble fractions was not constant with increasing plumieridine concentrations. As predicted by TMHMM, CHI2 possesses a transmembrane helix and may be responsible for chitinolytic activity detected in the insoluble cell fraction. This suggests CHI2 activity might not be inhibited by plumieridine. Additionally, molecular dynamics simulations indicated that CHI4 may not interact with plumieridine, as it happens with the other chitinases. Weaker or lack affinity suggests specificity with residues in the active site. Allosamidin is a known chitinase inhibitor isolated from Streptomyces spp. and is also regulates chitinase production in this species. Besides regulating chitinase production, allosamidin does not inhibit chitin-hydrolytic activity in chitinases from Streptomyces (Suzuki et al., 2006). In the other hand, plumieridine seems to reduce C. neoformans chitinolytic activity, but affects only CHI22 expression levels and may not be involved in chitinase transcription regulation. Several chitinases inhibitors have been reported (Schüttelkopf et al., 2010; Rao et al., 2005; Chen et al., 2014; Christy; Jayaprakash, 2017). Argifin is a potent chitinase inhibitor isolated from the fungi Gliocladium sp. and has an insecticide potential demonstrated in Periplaneta americana (Omura et al., 2000). Thus, the importance of chitinase in molecular processes broaden the applications for plumieridine. In this way, chitinase inhibitors can also be employed to relieve the symptoms of respiratory diseases, since chitinolytic activity and chitinase expression levels increases during pulmonary inflammations, aggravating it (Mazur et al., 2019; Létuvé et al., 2010). For instance, the widely studied chitinase inhibitor allosamidin was

45 observed to reduce asthma inflammatory process by reducing lymphocyte and eosinophil recruitment to mouse lungs (Zhu et al, 2004). Thus, virtual screening approach employed here to identify plumieridine’s target pointed at chitinase B1 from Aspergillus fumigatus complexed with the tripeptide VR0 (PDB ID 3CHE), which is derived from known inhibitor argifin (Andersen et al., 2008). Structural analysis reveals that the interactions in the C. neoformans complexes (chitinase-plumieridine) occurs with or near the residues of the conserved chitinase motif (DxDxE) (Figure S13) (Supplementary Material S13). Plumieridine’s initial position was the same as VR0 in A. fumigatus chitinase, however after dynamics simulation the inhibitor positions varied in the active site. Nonetheless, after simulations, the interactions between the motif DxDxE of C. neoformans’ chitinases (CHI2, CHI21, and CHI22) and plumierdine were still observed (Supplementary Material Figure S13). Notably, plumieridine binding is related to the GH18 superfamily chitinase inhibitor argifin (Andersen et al., 2008) suggesting that this molecule can be a general GH18 superfamily inhibitor. In order to test this hypothesis, in vitro and in silico experiments were conducted with supernatant and cell-free extracts of bacteria, insect and mouse cells. Inhibition of chitinolytic activity was observed in these assays, further supporting plumieridine as a chitinase inhibitor of broad spectrum and pointing for applications beyond cryptococcosis treatment. Contribution to the Field Statement Cryptococcosis is a fungal infection that affects people worldwide, especially immunocompromised patients (i. e. HIV/AIDS patients). The treatment available uses drugs that may present toxic effect for patients and lead to deadly fungal strains. This work tested the extract from seeds of a South America plant and identified the chemical compound responsible for the ability of killing the fungus. Through computational experiments, the molecule which the antifungal interacts was identified and simulations show the interaction effects key components for fungal reproduction. Research on new molecules capable of killing the fungus with less or none toxic effect for patients are urgently needed. We showed that the compound identified can cause fungal death by binding to a component of the assexual reproduction, this was observed through computational experiments and in the laboratory. Acknowledgments The authors would like to thank the Universidade Federal do Rio Grande do Norte and Universidade Federal do Rio Grande do Sul for providing the infrastructure and support and the Coordination for the Improvement of Higher Education Personnel (CAPES) for funding. Conflict of Interest Statement The authors have confirmed that there is no conflict of interest. Author Contributions Statement MH, JP, EG contributed for experimental design and results analysis. ES, VB contributed with seed extractraction, MIC assays, and chromatographic fractionation. ES, JC performed RT- qPCR. ES, RO contributed with chitinolytic activity assays. GL contributed with NMR analysis. AW contributed with macrophages cell culture and supernatant extraction. NS, ES with tenebrio’s chitinase extraction and manuscript redaction References Agustinho D. P. Nosanchuk, J. D. Functions of fungal melanins. In: Reference module in life sciences. (2017). Available from: http://www.sciencedirect.com/science/article/pii/B9780128096338120916. Agustinho, D. P., Miller, L. C., Li, L. X., & Doering, T. L. (2018). Peeling the onion: the outer layers of Cryptococcus neoformans. Mem Inst Oswaldo Cruz. 2018;113(7):e180040. doi: 10.1590/0074-02760180040.

Armenteros, J. J. A., Tsirigos, K. D., Sønderby, C. K., Petersen, T. N., Winther, O., Brunak, S., ... & Nielsen, H. (2019). SignalP 5.0 improves signal peptide predictions using deep neural networks. Nat Biotechnol 37, 420–423. https://doi.org/10.1038/s41587-019-0036-z. Atanasov, A. G., Waltenberger, B., Pferschy-Wenzig, E. M., Linder, T., Wawrosch, C., Uhrin, P., ... & Rollinger, J. M. (2015). Discovery and resupply of pharmacologically active plant- derived natural products: A review. Biotechnol Adv. 2015 Dec;33(8):1582-1614. doi: 10.1016/j.biotechadv.2015.08.001. Banks, I. R., Specht, C. A., Donlin, M. J., Gerik, K. J., Levitz, S.M., Lodge, J.K. A chitin synthase and its regulator protein are critical for chitosan production and growth of the fungal pathogen Cryptococcus neoformans. Eukaryot Cell. 4(11):1902-12. doi:10.1128/EC.4.11.1902–1912.2005. Chen, J., Varma, A., Diaz, M. R., Litvintseva, A. P., Wollenberg, K. K., & Kwon-Chung, K. J. (2008). Cryptococcus neoformans strains and infection in apparently immunocompetent patients, China. Emerg Infect Dis. 14(5):755-62. doi: 10.3201/eid1405.071312. Chen, L., Zhou, Y., Qu, M., Zhao, Y., & Yang, Q. (2014). Fully deacetylated chitooligosaccharides act as efficient glycoside hydrolase family 18 chitinase inhibitors. J. Biol. Chem., 289(25), 17932-17940. DOI: 10.1074/jbc.M114.564534 Christy, A., & Jayaprakash, K. Hitherto unobserved inhibition of insect chitinase enzyme by natural terpene. International Journal of Entomology Research. 2(3), 15-16. Coelho, C., & Casadevall, A. (2016). Cryptococcal therapies and drug targets: the old, the new and the promising. Cell Microbiol. 18(6):792-9. doi: 10.1111/cmi.12590. da Silva, D. L., Magalhães, T. F. F., Dos Santos, J. R. A., de Paula, T. P., Modolo, L. V., de Fátima, A., ... & de Resende‐Stoianoff, M. A. (2016). Curcumin enhances the activity of fluconazole against Cryptococcus gattii‐induced cryptococcosis infection in mice. J Appl Microbiol. 120(1):41-8. doi: 10.1111/jam.12966. Denning, D. W. (2003). Echinocandin antifungal drugs. Lancet. 4;362(9390):1142-51. https://doi.org/10.1016/S0140-6736(03)14472-8. Duo-Chuan, L. (2006). Review of fungal chitinases. Mycopathologia, 161(6), 345-360. https://doi.org/10.1007/s11046-006-0024-y Feldmesser, M., Kress, Y., Mednick, A., & Casadevall, A. (2000). The effect of the echinocandin analogue caspofungin on cell wall glucan synthesis by Cryptococcus neoformans. J Infect Dis. 182(6):1791-5. https://doi.org/10.1086/317614. Funkhouser, J. D., & Aronson, N. N. (2007). Chitinase family GH18: evolutionary insights from the genomic history of a diverse protein family. BMC Evol Biol. 2007 Jun 26;7:96. doi:10.1186/1471-2148-7-96. Gilbert, N. M., Donlin, M. J., Gerik, K. J., Specht, C. A., Djordjevic, J. T., Wilson, C. F., ... & Lodge, J. K. (2010). KRE genes are required for β‐1, 6‐glucan synthesis, maintenance of capsule architecture and cell wall protein anchoring in Cryptococcus neoformans. Mol Microbiol. 76(2):517-34. doi: 10.1111/j.1365-2958.2010.07119.x. Giles, S. S., Dagenais, T. R., Botts, M. R., Keller, N. P., & Hull, C. M. (2009). Elucidating the pathogenesis of spores from the human fungal pathogen Cryptococcus neoformans. Infect Immun. 77(8):3491-500. doi: 10.1128/IAI.00334-09. Grover, A. (2012). Plant chitinases: genetic diversity and physiological roles. Critical Reviews in Plant Sciences, 31(1):57-73. DOI: 10.1080/07352689.2011.616043 Han, B., Zhou, K., Li, Z., Sun, B., Ni, Q., Meng, X., ... & Zhou, C. (2016). Characterization of the first fungal glycosyl hydrolase family 19 chitinase (NbchiA) from Nosema bombycis (Nb). J Eukaryot Microbiol. 2016 63(1):37-45. doi: 10.1111/jeu.12246. Howard, M. B., Ekborg, N. A., Weiner, R. M., & Hutcheson, S. W. (2003). Detection and characterization of chitinases and other chitin-modifying enzymes. J Ind Microbiol Biotechnol. 2003 Nov;30(11):627-35. https://doi.org/10.1007/s10295-003-0096-3

Junges, Â., Boldo, J. T., Souza, B. K., Guedes, R. L. M., Sbaraini, N., Kmetzsch, L., ... & Vainstein, M. H. (2014). Genomic analyses and transcriptional profiles of the glycoside hydrolase family 18 genes of the entomopathogenic fungus Metarhizium anisopliae. PLoS ONE 9(9): e107864. doi:10.1371/journal.pone.0107864. Krysan, D. J. (2015). Toward improved anti-cryptococcal drugs: novel molecules and repurposed drugs. Fungal Genet Biol. 78:93-8. doi: 10.1016/j.fgb.2014.12.001. Kwon-Chung KJ, Bennett JE, Wickes BL, et al. (2017). The case for adopting the “species complex” nomenclature for the etiologic agents of cryptococcosis. mSphere. 11;2(1). pii: e00357-16. doi: 10.1128/mSphere.00357-16. Liu, X., & Kokare, C. (2017). Microbial enzymes of use in industry. In Biotechnology of Microbial Enzymes (pp. 267-298). Academic Press. https://doi.org/10.1016/B978-0-12- 803725-6.00011-X Lombard, V., Golaconda Ramulu, H., Drula, E., Coutinho, P. M., & Henrissat, B. (2014). The carbohydrate-active enzymes database (CAZy) in 2013. Nucleic Acids Res. 2014 Jan;42(Database issue):D490-5. doi: 10.1093/nar/gkt1178. Lu, S., Wang, J., Chitsaz, F., Derbyshire, M. K., Geer, R. C., Gonzales, N. R., ... & Thanki, N. (2020). CDD/SPARCLE: the conserved domain database in 2020. Nucleic acids research, 48(D1), D265-D268. Maligie, M. A., & Selitrennikoff, C. P. (2005). Cryptococcus neoformans resistance to echinocandins:(1, 3) β-glucan synthase activity is sensitive to echinocandins. Antimicrobial agents and chemotherapy, 49(7), 2851-2856. doi:10.1128/AAC.49.7.2851–2856.2005. Martinez, L. R., Moussai, D., & Casadevall, A. (2004). Antibody to Cryptococcus neoformans glucuronoxylomannan inhibits the release of capsular antigen. Infect Immun. 2004 Jun;72(6):3674-9. DOI: 10.1128/IAI.72.6.3674–3679.2004. Möller, S., Croning, M. D., & Apweiler, R. (2001). Evaluation of methods for the prediction of membrane spanning regions. Bioinformatics, 17(7), 646-653. https://doi.org/10.1093/bioinformatics/17.7.646. Mourad, A., & Perfect, J. R. (2018a). The war on cryptococcosis: a review of the antifungal arsenal. Mem Inst Oswaldo Cruz. 113(7):e170391. doi: 10.1590/0074-02760170391. Mourad, A., & Perfect, J. R. (2018b). Present and future therapy of cryptococcus infections. J Fungi (Basel). 4(3). pii: E79. doi: 10.3390/jof4030079. Oyeleye, A., & Normi, Y. M. (2018). Chitinase: diversity, limitations, and trends in engineering for suitable applications. Biosci Rep. 29;38(4). pii: BSR2018032300. doi: 10.1042/BSR20180323. Pappas, P. G. (2010). Cryptococcosis in the developing world: An elephant in the parlor. Clin. Infect. Dis. 50(3): 345–346. https://doi.org/10.1086/649862. Park, B. J., Wannemuehler, K. A., Marston, B. J., Govender, N., Pappas, P. G., & Chiller, T. M. (2009). Estimation of the current global burden of cryptococcal meningitis among persons living with HIV/AIDS. AIDS. 20;23(4):525-30. doi: 10.1097/QAD.0b013e328322ffac. Perfect, J. R., Dismukes, W. E., Dromer, F., Goldman, D. L., Graybill, J. R., Hamill, R. J., ... & Pappas, P. G. (2010). Clinical practice guidelines for the management of cryptococcal disease: 2010 update by the Infectious Diseases Society of America. Clin Infect Dis. 1;50(3):291-322. doi: 10.1086/649858. Poley, M., Koubek, R., Walsh, L., & McGillen, B. (2019). Cryptococcal Meningitis in an Apparent Immunocompetent Patient. J Investig Med High Impact Case Rep. 7:2324709619834578. doi: 10.1177/2324709619834578. Prakash, N. U., Jayanthi, M., Sabarinathan, R., Kangueane, P., Mathew, L., & Sekar, K. (2010). Evolution, homology conservation, and identification of unique sequence signatures in GH19 family chitinases. J Mol Evol. 70(5):466-78. doi: 10.1007/s00239-010-9345-z.

Rajasingham, R., Smith, R. M., Park, B. J., Jarvis, J. N., Govender, N. P., Chiller, T. M., ... & Boulware, D. R. (2017). Global burden of disease of HIV-associated cryptococcal meningitis: an updated analysis. Lancet Infect Dis. 2017 Aug;17(8):873-881. doi: 10.1016/S1473- 3099(17)30243-8. Rao, F. V., Houston, D. R., Boot, R. G., Aerts, J. M., Hodkinson, M., Adams, D. J., ... & van Aalten, D. M. (2005a). Specificity and affinity of natural product cyclopentapeptide inhibitors against A. fumigatus, human, and bacterial chitinases. Chemistry & biology, 12(1), 65-76. Rao, F. V., Andersen, O. A., Vora, K. A., DeMartino, J. A., & Van Aalten, D. M. (2005). Methylxanthine drugs are chitinase inhibitors: investigation of inhibition and binding modes. Chemistry & biology, 12(9), 973-980. Rathore, A. S., & Gupta, R. D. (2015). Chitinases from bacteria to human: properties, applications, and future perspectives. Enzyme Res. 2015;2015:791907. doi: 10.1155/2015/791907. Sahai, A. S., & Manocha, M. S. (1993). Chitinases of fungi and plants: their involvement in morphogenesis and host—parasite interaction. FEMS Microbiol Rev. 11(4), 317-338. DOI: 10.1016/0168-6445(93)90004-S. Santos-Gandelman, J., & Machado-Silva, A. (2019). Drug development for cryptococcosis treatment: what can patents tell us?. Mem. Inst. Oswaldo Cruz, 114(1). http://dx.doi.org/10.1590/0074-02760180391. Sloan, D. J., & Parris, V. (2014). Cryptococcal meningitis: epidemiology and therapeutic options. Clin Epidemiol. 13;6:169-82. doi: 10.2147/CLEP.S38850. Srikanta, D., Santiago‐Tirado, F. H., & Doering, T. L. (2014). Cryptococcus neoformans: historical curiosity to modern pathogen. Yeast. 31(2):47-60. doi: 10.1002/yea.2997. Stone, N. R., Rhodes, J., Fisher, M. C., Mfinanga, S., Kivuyo, S., Rugemalila, J., ... & Harrison, T. S. (2019). Dynamic ploidy changes drive fluconazole resistance in human cryptococcal meningitis. J Clin Invest. 129(3):999-1014. doi: 10.1172/JCI124516. Suchitha, S., Sheeladevi, C. S., Sunila, R., & Manjunath, G. V. (2012). Disseminated cryptococcosis in an immunocompetent patient: a case report. Case reports in pathology, 2012. https://doi.org/10.1155/2012/652351. Teixeira, A. P. D. C., Nóbrega, R. D. O., Lima, E. D. O., Araújo, W. D. O., & Lima, I. D. O. (2018). Antifungal activity study of the monoterpene thymol against Cryptococcus neoformans. Natural product research, 1-4. https://doi.org/10.1080/14786419.2018.1547296 Vaaje-Kolstad, G., Vasella, A., Peter, M. G., Netter, C., Houston, D. R., Westereng, B., ... & van Aalten, D. M. (2004). Interactions of a family 18 chitinase with the designed inhibitor HM508 and its degradation product, chitobiono-δ-lactone. Journal of Biological Chemistry, 279(5), 3612-3619. doi: 10.1074/jbc.M310057200 Valli, M., Pivatto, M., Danuello, A., Castro-Gamboa, I., Silva, D. H. S., Cavalheiro, A. J., ... & Bolzani, V. D. S. (2012). Tropical biodiversity: has it been a potential source of secondary metabolites useful for medicinal chemistry?. Química Nova, 35(11), 2278-2287. http://dx.doi.org/10.1590/S0100-40422012001100036. Velagapudi, R., Hsueh, Y. P., Geunes-Boyer, S., Wright, J. R., & Heitman, J. (2009). Spores as infectious propagules of Cryptococcus neoformans. Infect Immun. 77(10):4345-55. doi: 10.1128/IAI.00542-09.

Supplementary Material

1. Supplementary Figures and Tables

Figure S1. Graphical abstract presenting experiments involved in this study. The arrow direction indicates the flow of the investigation, since the plumieridine extraction from Allamanda polyantha seeds to the target identification, and the in vitro experiments. Table S1. Cryptococcus neoformans chitinases sequence information. This table is the result of sequence analysis with SignalP and TMHMM, Conserved Domain Database for prediction of signal peptides, transmembrane helices and identification of conserved domains, respectively. Together with results found by Baker and collaborators (2009) in regard to N- terminal sequence and serotype of C. neoformans chitinases.

Chitinase Broad Transmemb Conserved Signal N-terminal (gene) Institute rane helix domain peptide signal serotype A sequence CHI2 CNAG_0341 Yes GH18 No None 2 chitinase-like domain (cl10447) CHI21 CNAG_0259 No Glyco hydro Yes MHFVGSTT 8 18 domain LFVILTAL (cl23725) AVRSA CHI22 CNAG_0424 No Glyco hydro Yes MFLSTPAV 5 18 domain LSFVLLLA (cl23725) SQSSAQ CHI4 CNAG_0235 No GH18 Yes MYCTLAT 1 chitinase like LSLLALAE domain A (cl10447)

Adapted from Baker et al., 2009.

Table S2. List of predicted plumieridine targets based on Aspergillus fumigatus protein structures available on Protein Data Bank. Virtual screening was performed through a ligand-based approach and using pharmACOphore to compare ligand’s similarity. After the screening results were filtered for the fungus A. fumigatus and protein name, PDB ID number and ligand complexed in the structures identified.

Protein PDB ID Ligand Chitinase 3CHD Dipeptide (WRG) Farnesyltransferase 4LNG Farnesyldiphosphate and tipifarnib Chitinase 2A3C Pentoxifylline Farnesyltransferase 4LNB Farnesyldiphosphate and ethylenediamine Chitinase 1W9V Cyclopentapeptide (argifin) Chitinase 2IUZ C2-dicaffeine Chitinase 3CHC Monopeptide (ZRG) Chitinase 4TX6 3-(2-methoxyphenyl)-6- methyl[1,2]oxazolo[5,4-d] pyrimidin-4(5H)-one Chitinase 2XTK Acetazolamide Chitinase 3CHE Tripeptide (VR0) Chitinase 3CHF Tetrapeptide (VR0) Chitinase 2A3A Theophylline Chitinase 3CH9 Dimethylguanylurea Fucose binding lectin 4D52 L-Galactopyranose FtmOx1 4Y5S 2-Oxoglutaric acid Chitinase 2A3B Caffeine Chitinase 2XUC 1-methyl-3-(N- methylcarbamimidoyl)urea Chitinase 1W9U Cyclopentapeptide (argadin) Fucose binding lectin 4C1Y B-methylfucoside

Figure S2. Plumieridine’s docked orientations in the active site of Cryptococcus neoformans chitinases. Orientations of plumieridine docked in the active site are indicated by the arrow direction. (A) inward orientation (red) in the active site; (B) outward orientation (yellow).

Figure S3. Cryptococcus neoformans chitinase molecular models. (A) Model for CHI2; (B) model for CHI21; (C) CHI22 model; and (D) CHI4 molecular model. All models were created with the Swiss model and evaluated with Swiss model Evaluation tool.

Figure S4. Cryptococcus neoformans chitinases sequences alignment. Alignment of chitinase sequences CHI2, CHI22, CHI21, and CHI4 with a focus for the Glycoside Hydrolase 18 superfamily DxDxE conserved motif. Figure S5. Hydrogen Magnetic Nuclear Resonance spectra from a plumieridine-rich chromatographic fraction extracted from Allamanda polyantha. NMR analysis was used to identify the compounds in the fraction with antifungal activity. Through NMR spectra is only possible to identify plumieridine’s structure in this fraction.

Figure S6. Carbon Magnetic Nuclear Resonance spectra from a plumieridine-rich chromatographic fraction extracted from Allamanda polyantha. NMR analysis was used to identify the compounds in the fraction with antifungal activity. Through NMR spectra is only possible to identify plumieridine’s structure in this fraction.

Figure S7. Trichoderma harzianum as the positive control for chitinolytic activity assays. (A) superposition of T. harzianum chitinase (PDB 6EPB) with C. neoformans CHI21 chitinase. 6EPB is shown in blue and CHI21 in golden. Superposition has 0.964 Å; (B) inhibition of

54 chitinolytic activity by plumieridine on a solution of Lysing Enzymes from Trichoderma harzianum, which contains chitinases. Through this assay, this solution was chosen as positive control for the inhibitory assays.

Figure S8. Cryptococcus neoformans inhibition of chitinolytic activity by plumieridine. (A) Inhibitory assay for the soluble cell fraction of fugal grown on YPD; (B) and YPGlcNAc (C) inhibitory assay for the insoluble cell fraction of fugal grown on YPD; (D) and YPGlcNAc. Letters above bars indicate statistical difference between different concentrations.

Figure S9. Cryptococcus neoformans relative expression levels. (A) Relative expression levels of CHI2 from cells grown on YPD in the presence and absence of plumieridine; (B) and on YPGlcNAc; (C) expression levels for CHI4 from cryptocococi cells grown on YPD; (D) and on YPGlcNAc; (E) CHI21 expression levels of cells incubated on YPD; (F) and on YPGlcNAc. Statistical significance was not observed for CHI4 expression levels under different treatments.

Figure S10. Details of Cryptococccus neoformans cells treated with 300 µg/mL of plumieridine. White arrow in (A) shows incomplete mother-daughter separation. Dotted arrows in (B) points to a cell with diffuse WGA staining pattern, while solid arrow points to a cell with polarized WGA staining pattern.

Figure S11. Superposition of murine’s chitinase like protein (Ym1) (PDB 1VF8) with human chitinase (PDB 1HKI). Murine chitinase is shown in blue and human chitinase in golden. Superposition has 0.692 Å.

Figure S12. Root mean standard deviation (RMSD) of chitinases backbones during 100 ps dynamics simulation. RMSD was performed with the two bound plumieridine orientations for each chitinase. (A) RMSD for CHI22-plumieridine complexes; (B) for CHI2- plumieridine complexes; (C) RMSD for CHI4-plumieridine complexes; (D) RMSD for CHI21- plumieridine complexes.

Figure S13. Interaction of plumieridine with residues of GH18 superfamily conserved motif after dynamics simulation. (A) Interaction of plumieridine with glutamic acid in CHI2 in the initial inward orientation; (B) plumieridine interaction with glutamic acid in 21 in the initial inward orientation; (C) interaction between plumieridine and aspartic acid of CHI22 in the starting outward orientation; (D) glutamic acid of CHI21 interacting with plumieridine in the initial outward orientation. Plumieridine is colored in red and residues of the DxDxE motif participating in the interaction are depicted in cyan.