JONAS HENRIQUE COSTA

ESTUDO METABÓLICO E DOS MECANISMOS DE PATOGENICIDADE DO

FUNGO Penicillium digitatum FRENTE AO SEU HOSPEDEIRO CITROS

Tese de Doutorado apresentada ao Instituto de Química da

Universidade Estadual de Campinas como parte dos
requisitos exigidos para a obtenção do título de Doutor em
Ciências.

Orientadora: Profa. Dra. Taicia Pacheco Fill

O arquivo digital corresponde à versão final da Tese defendida pelo aluno

Jonas Henrique Costa e orientada pela Profa. Dra. Taicia Pacheco Fill.

CAMPINAS
2021
 Ficha catalográfica
Universidade Estadual de Campinas
Biblioteca do Instituto de Química
Simone Luiz Alves - CRB 8/9094

Costa, Jonas Henrique, 1989-

C823e CosEstudo metabólico e dos mecanismos de patogenicidade do fungo
Penicillium digitatum frente ao seu hospedeiro citros / Jonas Henrique Costa. –
Campinas, SP : [s.n.], 2021.

CosOrientador: Taicia Pacheco Fill.

CosTese (doutorado) – Universidade Estadual de Campinas, Instituto de
Química.

Cos1. Fungos fitopatogênicos. 2. Penicillium digitatum. 3. Cítricos. 4.

Metabólitos. I. Fill, Taicia Pacheco, 1983-. II. Universidade Estadual de
Campinas. Instituto de Química. III. Título.

Informações para Biblioteca Digital

Título em outro idioma: A study of the metabolism and pathogenicity mechanisms of the
fungus Penicillium digitatum against citrus host
Palavras-chave em inglês:
Phytopathogenic fungi
Penicillium digitatum
Citrus
Metabolites
Área de concentração: Química Orgânica
Titulação: Doutor em Ciências
Banca examinadora:
Taicia Pacheco Fill [Orientador]
Fabio Cesar Gozzo
Mônica Tallarico Pupo
Edson Rodrigues Filho
Luciana Gonzaga de Oliveira
Data de defesa: 26-03-2021
Programa de Pós-Graduação: Química

Identificação e informações acadêmicas do(a) aluno(a)

- ORCID do autor: https://orcid.org/0000-0002-6415-3984
- Currículo Lattes do autor: http://lattes.cnpq.br/8735544113358429

Powered by TCPDF (www.tcpdf.org)

BANCA EXAMINADORA

Profa. Dra. Taicia Pacheco Fill (Orientadora)

Profa. Dra. Mônica Tallarico Pupo (Universidade de São Paulo)

Prof. Dr. Edson Rodrigues Filho (Universidade Federal de São Carlos)

Profa. Dra. Luciana Gonzaga de Oliveira (Universidade Estadual de Campinas)

Prof. Dr. Fabio Cesar Gozzo (Universidade Estadual de Campinas)

A Ata da defesa assinada pelos membros da Comissão Examinadora, consta no

SIGA/Sistema de Fluxo de Dissertação/Tese e na Secretaria do Programa da Unidade.

Este exemplar corresponde à redação

final da Tese de Doutorado defendida
pelo aluno JONAS HENRIQUE COSTA,
aprovada pela Comissão Julgadora em
26 de março de 2021.
 AGRADECIMENTOS

Aos meus pais Paulo e Sandra, e minha irmã Beatriz que sempre apoiaram
os meus estudos.
Ao meu border collie Safári que sempre me recebe em casa, após a
Unicamp, com a maior alegria.
Ao Jeff por todo o suporte emocional e auxílio no inglês.
Aos colegas do Labioquimi, em especial: Aline, Jaque, Dani, Laurinha,
Gabriel, Pedro Luis e Mayra. O doutorado não teria a menor graça sem vocês!
Obrigado pela amizade e por todos os rodízios de comida japonesa que levamos à
falência.
À minha orientadora Taicia por todos os ensinamentos, ajudas, conselhos
e suporte. Eu não teria conseguido ir tão longe se não fosse a atenção e dedicação
que você tem com seus alunos. Com a sua ajuda, meu pote está cheio de moedinhas!
Aos colegas Pedro, Amanda, Janaína, Isabela e Caio.
Ao Instituto de Química que me acolheu durante mais de seis anos.
À Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil
(CAPES) – Código de Financimento 001 – pelo auxílio financeiro.
À todos os funcionários e docentes do Instituto de Química e da Unicamp
que de alguma maneira contribuíram pela minha passagem na universidade.

MUITO OBRIGADO!!!

Jonas
 RESUMO

O Brasil é o maior produtor mundial de laranja. Infelizmente, as frutas

cítricas são acometidas por doenças causadas por fitopatógenos. Nos citros, a mais
destrutiva doença pós colheita é o bolor verde causado pelo fungo Penicillium
digitatum que é responsável por até 90 % das perdas dos frutos. Considerando a
relevância da citricultura para o Brasil e os danos causados por P. digitatum, esta tese
tem como objetivo estudar o metabolismo e os mecanismos de patogenicidade do
fungo P. digitatum. Desta forma, a introdução da tese apresenta uma revisão
bibliográfica à respeito de P. digitatum e seus mecanismos de infecção.
Nos capítulos I e II, estudos sobre o perfil metabólico de P. digitatum foram
feitos, em paralelo, com bioensaios para investigar o papel biológico dos principais
metabólitos secundários deste fitopatógeno. No capítulo I, o imageamento por
espectrometria de massas (IMS) foi aplicado para monitorar os metabólitos
secundários produzidos na superfície de laranjas infectadas com P. digitatum,
revelando uma acumulação de alcalóides triptoquialaninas na superfície dos frutos. A
triptoquialanina A foi submetida a bioensaios inseticidas que revelaram sua alta
toxicidade contra Aedes aegypti, sugerindo uma importante ação inseticida durante o
apodrecimento dos frutos.
No capítulo II, as triptoquialaninas foram avaliadas em relação à
fitotoxicidade. A triptoquialanina A inibiu a germinação das sementes de Citrus
sinensis. Considerando as bioatividades apresentas pelas triptoquialaninas, o
mecanismo de transporte no qual esses alcaloides são exportados para o meio
extracelular foi investigado e, pela primeira vez, vesículas extracelulares (VEs)
secretadas por P. digitatum foram detectadas in vitro e in vivo. O conteúdo das VEs
foi estudado e observou-se que as triptoquialaninas e outros metabólitos secundários
são transportados através das VEs durante o processo de infecção. A produção de
VEs por P. digitatum abre novos caminhos na elucidação dos fatores de virulência
utilizados por este fitopatógeno.
No capítulo III, as estratégias de virulência de fitopatógenos de citros foram
investigadas, em relação às atividades enzimáticas, através de ensaios de biocatálise.
Flavonoides cítricos, compostos envolvidos na defesa da planta, foram utilizados
como substrato. Os ensaios revelaram diferentes perfis de hidrólise entre os
fitopatógenos. Além disso, novos flavonoides hidroxilados produzidos pela planta
foram detectados em resposta à infecção fúngica.
Por fim, no capítulo IV, a estratégia de co-cultivo envolvendo fungos
patógenos de citros (P. digitatum e Penicillium citrinum) foi aplicada visando a busca
de métodos mais seguros para o controle do bolor verde através do uso de produtos
naturais. O IMS revelou uma guerra química na qual dois tetrapeptídeos,
deoxicitrinadina A, citrinadina A, chrisogenamida A e triptoquialaninas foram
detectados na zona de confronto da co-cultura. Esses metabólitos foram isolados e
caracterizados por análises de espectrometria de massas e ressonância magnética
nuclear. Ensaios de atividade antimicrobiana utilizando os metabólitos isolados da
zona de confronto confirmaram as atividades antifúngicas. O crescimento de P.
digitatum foi inibido em meio de cultura suplementado com compostos produzidos por
P. citrinum.
 ABSTRACT

Brazil is the world's largest orange producer. Unfortunately, citrus fruits are
affected by diseases caused by phytopathogens. In citrus, the most destructive post-
harvest disease is the green mold caused by the fungus Penicillium digitatum, which
is responsible for up to 90% of fruit losses. Considering the relevance of citriculture to
Brazil and the damage caused by P. digitatum, this thesis aims to estudy the
metabolism and pathogenic mechanisms of the fungus P. digitatum. Thus, the
introduction provides a bibliographic review of P. digitatum and its infection
mechanisms.
In chapters I and II, studies on the metabolic profile of P. digitatum were
carried out, in parallel, with bioassays to investigate the biological role of the main
secondary metabolites of this phytopathogen. In chapter I, imaging mass spectrometry
(IMS) was applied to monitor the secondary metabolites produced on the surface of
oranges infected with green mold, revealing an accumulation of tryptoquialanines
alkaloids on the fruit surface. Tryptoquialanine A was submitted to insecticidal
bioassays that revealed its high toxicity against Aedes aegypti, suggesting an
important insecticidal action during the fruit decay.
In chapter II, tryptoquialanines were evaluated regarding to phytotoxicity.
Tryptoquialanine A inhibited the germination of Citrus sinensis seeds. Considering the
bioactivities presented by tryptoquialanines, the transport mechanism in which these
alkaloids are exported to the extracellular environment was investigated and, for the
first time, extracellular vesicles (EVs) secreted by P. digitatum were detected in vitro
and in vivo. EVs cargo were studied and it has been observed that tryptoquialanines
and other secondary metabolites are transported through EVs during the infection
process. The production of EVs by P. digitatum is a open field that could help to
elucidate the virulence factors used by this phytopathogen.
In chapter III, the virulence strategies of citrus phytopathogens were
investigated, regarding to enzymatic activities, through biocatalysis assays. Citrus
flavonoids, compounds involved in plant defense, were used as substrate. The assays
revealed different hydrolysis profiles between the phytopathogens. In addition, novel
hydroxylated flavonoids produced by the plant have been detected in response to
fungal infection.
 Finally, in Chapter IV, the co-cultive strategy involving citrus pathogenic
fungi (P. digitatum and Penicillium citrinum was applied in order to search for safer
methods to control the green mold disease through the use of natural products. IMS
has revealed a chemical war in which two tetrapeptides, deoxycitrinadin A, citrinadin
A, chrysogenamide A and tryptoquialanines are produced in the co-culture
confrontation zone. This metabolites were purified and characterized by mass
spectrometry and nuclear magnetic resonance analysis. Antimicrobial activity assays
using the metabolites purified from the co-culture confrontation zone confirmed the
antifungal activities. P. digitatum’ growth was inhibited in culture medium
supplemented with compounds produced by P. citrinum.
 SUMÁRIO

1 INTRODUÇÃO
1.1 A relevância dos produtos naturais ......................................................12
1.2 Penicillium digitatum e a doença do bolor verde...................................14
1.3 Artigo de revisão: “Penicillium digitatum infection mechanisms in citrus:
What do we know so far?”……………….....................................................16

2 OBJETIVOS............................................................................................................27

3 CAPÍTULO I: Monitoramento da produção de alcalóides indólicos por

Penicillium digitatum durante o processo de infecção em citros utilizando
imagem de espectrometria de massas e rede molecular
3.1 Resumo …………………..………………………………………………...28
3.2 Artigo: “Monitoring indole alkaloid production by Penicillium digitatum
during infection process in citrus by Mass Spectrometry Imaging and
molecular networking” …………………………………………………………29
3.3 Informações suplementares do capítulo I ............................................36

4 CAPÍTULO II: Triptoquialaninas fitotóxicas produzidas in vivo por Penicillium

digitatum são exportadas em vesículas extracelulares
4.1 Resumo ………………………..…………………………………………...51
4.2 Artigo: “Phytotoxic tryptoquialanines produced in vivo by Penicillium
digitatum are exported in extracellular vesicles” ………………………...….52
4.3 Informações suplementares do capítulo II ………………….................68

5 CAPÍTULO III: Explorando a interação entre flavonóides cítricos e fungos

fitopatogênicos através de atividades enzimáticas
5.1 Resumo...............................................................................................90
5.2 Artigo: “Exploring the interaction between citrus flavonoids and
phytopathogenic fungi through enzymatic activities” ………………....……91
5.3 Informações suplementares do capítulo III .........................................99
6 CAPÍTULO IV: Potencial antifúngico de metabólitos secundários envolvidos
na interação entre patógenos cítricos
6.1 Resumo …………….……………………….………………………….…113
6.2 Artigo: “Antifungal potential of secondary metabolites involved in the
interaction between citrus pathogens” ……………………………………..114
6.3 Informações suplementares do capítulo IV ........................................125

7 DISCUSSÃO.........................................................................................................151

8 CONCLUSÕES E PERSPECTIVAS......................................................................155

9 REFERÊNCIAS BIBLIOGRÁFICAS ....................................................................157

10 ANEXOS
10.1 Licença de reprodução do artigo de revisão e dos capítulos I e III....161
10.2 Licença de reprodução do capítulo II e IV .......................................162
10.3 Autorização da Comissão Interna de Biossegurança (CIBio-IQ) .....163
10.4 Autorização de acesso ao patrimônio genético (SISGEN) ...............164
10.5 Declaração de tese em formato alternativo .....................................165
 12

1. INTRODUÇÃO

1.1 A relevância dos produtos naturais

Os produtos naturais são utilizados pela humanidade há séculos, de forma
que a utilização de ervas e folhas na cura de doenças podem ser ditas como uma das
primeiras formas de aplicação destes compostos1. Assim, a química dos produtos
naturais é a subárea mais antiga da química orgânica2.
Os produtos naturais podem ser definidos como compostos químicos
produzidos por um ser vivo encontrado na natureza (animais, plantas e
microrganismos) e que não estão envolvidos no metabolismo primário, sendo
codificadas geneticamente e produzidas por vias metabólicas secundárias3. Essas
moléculas possuem uma diversidade de estruturas químicas 4 e, além disso, são
reconhecidas como fontes de inspiração para a química, biologia e especialmente
para medicina, onde vêm há tempos ajudando a combater doenças, revolucionando a
área médica4,5.
Até os anos de 1960, as pesquisas nessa área mantinham um ritmo lento,
com a descoberta de apenas poucos compostos por ano 2,3. Porém, com o
desenvolvimento das técnicas de separação e detecção como High Performance
Liquid Chromatography (HPLC) e Ressonância Magnética Nuclear (RMN), o
isolamento e elucidação de compostos naturais foi aprimorado, permitindo um melhor
conhecimento sobre o metabolismo secundário de organismos vivos e levando a um
aumento no número de novas moléculas reportadas anualmente 2,3,6. A Figura 1
mostra que a descoberta de novos produtos naturais aumentou significamente a partir
de 1960, partindo de poucos compostos em 1940 e chegando a uma média de 1600
compostos por ano nas últimas duas décadas3. Entretanto, percebe-se que desde
1990 o número de novas estruturas reportadas se mantém estável e a taxa de
estruturas únicas representa uma porcentagem decrescente do número total de
compostos isolados3. Portanto, priorizar a descoberta de moléculas únicas vêm se
tornado uma questão central na área de produtos naturais3.
 13

Figura 1. Número de novos produtos naturais (PN) reportados por ano e taxa de isolamento
de compostos novos com estruturas únicas3. Adaptado de Pye et al. (2017).

A química dos produtos naturais possui impacto em vários campos do

conhecimento e inclusive na atividade econômica6. Diversos fármacos aprovados
originaram-se de produtos naturais, incluindo compostos inalterados, quimicamente
modificados ou sintéticos que utilizam compostos naturais como modelo4. Como
exemplo de agentes quimioterapêuticos, temos o paclitaxel, comercialmente
conhecido como taxol e extraído da casca da árvore teixo-do-pacífico (Taxus
brevifolia), e a vimblastina e vincristina, extraídas da pequena planta vinca-de-
madagascar (Catharrantus roseu) 1,7.

Figura 2. Estruturas de compostos naturais utilizados no tratamento contra o câncer:

paclitaxel (1), vimblastina (2) e vincristina (3).

Além do impacto econômico e da importância na indústria farmacêutica, os

produtos naturais ainda são utilizados como produtos consumíveis (pigmentos,
temperos, óleos), produtos técnicos, fibras e polímeros (como a celulose),
surfactantes, pesticidas e larvicidas naturais e outros5,6.
A pesquisa em produtos naturais de plantas ainda é a principal, enquanto
os microrganismos, principalmente os fungos, são a segunda fonte mais explorada
por pesquisadores brasileiros2. Nas últimas décadas, os microrganismos vêm
recebendo atenção das indústrias e dos pesquisadores da área de produtos naturais1,
 14

uma vez que um único microrganismo pode produzir diversos metabólitos

secundários, chegando a um valor até maior que 50 compostos5.
Dentre os compostos biologicamente ativos, 60 % são oriundos de plantas
e grande parte do restante provém de microrganismos. Nas drogas medicinais os
microrganismos também desempenham papel importante5, 23 mil compostos ativos
são conhecidos por terem origem microbiana, sendo que 42 % deles tem sua origem
em fungos e 32% em bactérias filamentosas5. Dos medicamentos provenientes de
microrganismos destacam-se os antibióticos, com cerca de 150 opções no mercado5.
Apesar do crescente investimento na pesquisa de produtos naturais de
microrganismos, estima-se que 95 a 99 % dos organismos vivos existentes não foram
cultivados em laboratório, de forma que apenas 1 % das espécies de bactérias e 5 %
das espécies de fungos tiveram seus metabólitos secundários estudados 4,5.
Além da baixa diversidade de microrganismos estudados, outra lacuna na
área de produtos naturais é a pouca informação sobre as atividades biológicas de
grande parte dos produtos naturais encontrados. Os desafios da química dos produtos
naturais vão além da elucidação estrutural e envolvem a compreensão da relação
entre a estrutura química e atividade biológica8. Há uma vasta oportunidade de
caracterizar as funções biológicas de produtos naturais complexos que ainda se
encontra totalmente inexplorada3.

1.2 Penicillium digitatum e a doença do bolor verde

As frutas cítricas possuem importante impacto na economia mundial, uma
vez que a citricultura é uma atividade bilionária e os citrinos possuem uma das maiores
produções globais quando comparados com outros frutos9, 10. Para o ano de 2019/20
somente a produção global de laranja está prevista para 47.5 milhões de toneladas11.
Por sua vez, a produção total de citros em 2019/20 está prevista para 94 milhões de
toneladas, incluindo limões, tangerinas e toranjas11.
Entre os maiores produtores de citros estão Brasil, China e Estados
Unidos11. Dentre os três países, o Brasil desponta como o maior produtor com uma
previsão de produção mundial de laranja e suco de laranja, respectivamente, de 57.6
e 31.8 % para o ano de 202011 (Figura 3). A citricultura brasileira é responsável por
230.000 empregos diretos e indiretos, e ainda movimenta, anualmente, 174 milhões
de doláres12, sendo uma atividade que envolve a produção dos citros nas fazendas, o
processamento dos frutos nas indústrias e a distribuição de vários produtos cítricos 10.
 15

Figura 3. Previsão 2020 para a produção mundial de laranjas e suco de laranja11. O Brasil
desponta como o maior produtor mundial em ambos os casos.

Apesar de sua importância econômica, os citros são afetados por doenças

pós-colheitas que afetam a qualidade das frutas e causam a perda de até 50 % dos
frutos, trazendo prejuízos e grande impacto econômico13,14. Nos citros, a doença mais
agressiva é o bolor verde causado pelo fungo Penicillium digitatum (Figura 4),
responsável por até 90 % das perdas no período de pós-colheita9,15.
Atualmente, o controle de P. digitatum e de outras doenças pós-colheita em
citros é feito através da aplicação em massa de fungicidas sintéticos como, prochloraz,
imazalil e pyrimethanil16. Entretanto, o uso repetido de fungicidas tem levado ao
surgimento de linhagens resistentes de P. digitatum, representando um obstáculo a
conservação dos frutos17. Além disso, a utilização de fungicidas causa preocupações,
devido a toxicidade dos produtos utilizados, em relação ao meio ambiente e à saúde
humana16.

Figura 4. Bolor verde causado por P. digitatum em citros. A coloração esverdeada é

característica da doença e provém da formação de esporos.

Várias abordagens vêm sendo propostas para substituir os fungicidas

sintéticos utilizados no controle de doenças pós-colheita de citros. Uma das
estratégias mais recentes é baseada na compreensão da função das fitotoxinas
produzidas por fungos patógenos em seu contexto biológico e os respectivos
 16

mecanismos de ação dessas substâncias18. A interação patógeno-hospedeiro é um

campo de batalhas de substâncias e alguns metabólitos secundários produzidos por
fungos fitopatógenos têm sido associados à patogenicidade frente aos hospedeiros 19.
Entretanto, apesar do genoma de P. digitatum ter sido sequenciado e indicar uma
grande quantidade de informação em relação a produção de metabólitos
secundários20, poucos trabalhos na literatura buscam correlacionar esses metabólitos
a um papel biológico no processo de infecção.
Assim, a introdução desta tese traz uma revisão bibliográfica sobre P.
digitatum, tendo em vista a grande quantidade de informações que são encontradas
na literatura a respeito deste fitopatógeno. O artigo de revisão apresenta as principais
caractéristicas da doença do bolor verde, os mecanismos de virulência já estudados,
os metabólitos secundários reportados, estudos de biologia molecular contendo genes
essenciais para a patogenicidade e o que ainda é desconhecido para esse
fitopatógeno na literatura.

1.3 Artigo de revisão: “Penicillium digitatum infection mechanisms in citrus:

What do we know so far?”

Parte do conteúdo desta introdução é composto pelo artigo intitulado “Penicillium

digitatum infection mechanisms in citrus: What do we know so far?”,
publicado no periódico Fungal Biology. A reprodução deste documento pelos
autores, para fins não comerciais, é autorizada pela editora Elsevier e não
necessita de permissão escrita (anexo 10.1).

Referência: Costa, J.H., Bazioli, J.M., Pontes, J.G.M., Fill, T.P. 2019. Penicillium
digitatum infection mechanisms in citrus: What do we know so far? Fungal Biology,
v. 123, p. 584-593. doi: 10.1016/j.funbio.2019.05.004

Versão final publicada disponível em:

https://www.sciencedirect.com/science/article/pii/S1878614619300601
 17

Contents lists available at ScienceDirect

Fungal Biology
journal homepage: www.elsevier.com/locate/funbio

Penicillium digitatum infection mechanisms in citrus: What do we

know so far?
Jonas Henrique Costa a, Jaqueline Moraes Bazioli a, b,
~o Guilherme de Moraes Pontes a, **, Taícia Pacheco Fill a, *
Joa
a
Institute of Chemistry, Universidade Estadual de Campinas, CP 6154, 13083-970 Campinas, SP, Brazil
b
Faculty of Pharmaceutical Sciences, Universidade Estadual de Campinas, 13083-859 Campinas, SP, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: Penicillium digitatum is the major source of postharvest decay in citrus fruits worldwide. This fungus
Received 2 February 2019 shows a limited host range, being able to infect mainly mature fruit belonging to the Rutaceae family.
Received in revised form This highly specific host interaction has attracted the interest of the scientific community. Researchers
26 April 2019
have investigated the chemical interactions and specialized virulence strategies that facilitate this fun-
Accepted 4 May 2019
Available online 13 May 2019
gus's fruit colonization, thereby leading to a successful citrus infection. There are several factors that
mediate and affect the interaction between P. digitatum and its host citrus, including hydrogen peroxide
Corresponding Editor: Gustavo Henrique modulation, secretion of organic acids and consequently pH control, and other strategies described here.
Goldman The recently achieved sequencing of the complete P. digitatum genome opened up new possibilities for
exploration of the virulence factors related to the host-pathogen interaction. Through such techniques as
Keywords: RNAseq, RT-PCR and targeted gene knockout mediated by Agrobacterium tumefaciens, important genes
Green mold disease involved in the fungal infection process in citrus have been reported, helping to elucidate the molecular
Postharvest pathogen mechanisms, metabolites and genetic components that are involved in the pathogenicity of P. digitatum.
Secondary metabolites
Understanding the infection process and fungal strategies represents an important step in developing
Virulence factors
ways to protect citrus from P. digitatum infection, possibly leading to more productive citriculture.
© 2019 British Mycological Society. Published by Elsevier Ltd. All rights reserved.

1. Introduction Various postharvest pathogens are reported to cause significant

Penicillium digitatum is a species within the Ascomycota division
of fungi with considerable importance to the environment as well
as to the food industry (Frisvad and Samson, 2004). This species is
the major source of postharvest decay in citrus fruits worldwide,
followed by Penicillium italicum (Poppe et al., 2003). In addition,
P. digitatum was surprisingly reported as a plum postharvest
pathogen, but this host-pathogen interaction has not been thor-
oughly elucidated to date (Louw and Korsten, 2019).
Economically, citrus fruits have the largest production in the
world compared to other fruits, with a global citrus production of
approximately 98.3 million tons, including oranges, tangerines,
lemons and grapefruit (Citrus: World Markets and Trade), indi-
cating their global economic importance (Molto  et al., 2010).
* Corresponding author.
** Corresponding author.
E-mail addresses: joao.pontes@iqm.unicamp.br (J.G. de Moraes Pontes), taicia@
iqm.unicamp.br (T.P. Fill).
losses at different storage stages after harvest, accounting for nearly
50 % of the wastage in citrus fruits (Ladanyia, 2010). In particular,
P. digitatum is one of the most severe postharvest pathogens,
causing green mold disease (Ghooshkhaneh et al., 2018), which
results in damage to citriculture and commerce. In arid zones and
tropical subclimates, P. digitatum has been reported to contribute to
90 % of the total post harvest losses in citriculture (Macarisin et al.,
2007).
P. digitatum is able to invade and infect the fruit through wounds
that can originate from environmental factors, such as wind, hail
and insects, or during the harvest process, transport and subse-
quent treatments (Perez et al., 2017). This fungus spreads on a
number of skin oil glands through pores and wounds
(Ghooshkhaneh et al., 2018), in which nutrients are available to
stimulate spore germination. The initiation of the infection area
appears as a soft watery spot (sometimes referred to as clear rot) on
the fruit peel, and if suitable temperature and conditions are
available, it becomes a white mycelium that later turns an olive

https://doi.org/10.1016/j.funbio.2019.05.004
1878-6146/© 2019 British Mycological Society. Published by Elsevier Ltd. All rights reserved.

J.H. Costa et al. / Fungal Biology 123 (2019) 584e593
color with spore production (Vu et al., 2018; Ismail and Zhang,
2004).
Currently, the postharvest control is carried out by the mass
application of fungicides in commercialized fruits. A number of
synthetic fungicides, such as prochloraz, imazalil, and pyrimethanil,
have been used worldwide as postharvest treatments to control the
pathogen in citrus fruits (Hao et al., 2011). Nevertheless, the
repeated use of certain fungicides has led to the appearance of
fungicide-resistant populations of P. digitatum, representing a sig-
nificant obstacle to postharvest conservation (Kanetis et al., 2010).
In addition to the serious implications for product toxicity, envi-
ronmental risk and low consumer confidence (Hao et al., 2011).
Alternative strategies to control P. digitatum infection have been
investigated. Recently, Lafuente et al. (2018) verified the effect of
light-emitting diode (LED) blue light (LBL) at a 450-nm wavelength
reducing the viability and the capacity of the infection by
P. digitatum in citrus fruits. The application of noncontinuous LBL is
a promising alternative to decrease the capacity of postharvest
decay since it is responsible for anomalous fungal spore morphol-
ogies and darkening of the cytoplasm. In addition, a significant
decrease in ethylene production in plates that were exposed to the
continuous LBL in relation to plates under darkness was observed
over eight days.
Studies concerning the citrusePenicillium system have mainly
focused on new alternatives for controlling the disease or on the
analysis of the mechanisms involved in induced resistance to fungal
infection. However, the recently sequenced complete genome
opened up new possibilities for investigating the virulence factors
related to the host-pathogen interaction. Here, we describe the
known mechanisms and important genes related to cit-
rusePenicillium interactions. Understanding the infection process
and fungal strategies may represent an important step in devel-
oping ways to protect citrus from P. digitatum infections, leading to
more productive citriculture worldwide.
2. P. digitatum e citrus interaction
There are several factors that mediate and affect the interaction
between P. digitatum and its host during infection. In this section,
we briefly describe the molecules and variables that directly or
indirectly influence the infection process.
2.1. Hydrogen peroxide production as a plant response
Hydrogen peroxide production by the host has been described
as an important defense mechanism, which is bypassed by
P. digitatum due to increased catalase production. Hydrogen
peroxide is mostly an initial consequence of plant defense in
response to pathogen infection, functioning as a signaling molecule
(Levine et al., 1994; Mellersh et al., 2002; Qin et al., 2011), and it has
been reported to play a role as a signal in the induction of systemic
acquired resistance (SAR) in plants (Neuencchwander et al., 1995;
Hunt et al., 1996; Conrath, 2006). Hydrogen peroxide also partici-
pates in lignification processes of the cell wall, oxidative cross-
linking of proteins (Olson and Varner, 1993; Brisson et al., 1994),
phytoalexin biosynthesis (Apostol et al., 1989) and induction of host
defense genes (Nathues et al., 2004), inhibiting pathogen growth
(Lu and Higgins, 1999; Ku zniak and Urbanek, 2000; García-Olmedo
et al., 2001).
Macarisin et al. (2007) performed a detailed study on the level of
hydrogen peroxide during the infection process of citrus fruits with
P. digitatum (compatible interactions) and with Penicillium expan-
sum (nonhost interactions). The authors presented clear evidence
that P. digitatum is able to avoid the H2O2-oxidative burst during the
first 25 h after inoculation in host cells, while P. expansum triggers a
18
considerable accumulation of H2O2 in the host (Macarisin et al.,
2007). In addition, the authors correlate the effect of organic
acids and low pH buffer with H2O2 host production based on the
evaluation of whether the accumulation of certain organic acids
might result in inhibition of the oxidative burst. Treatment with
citric acid, oxalic acid and ascorbic acid contributed the most to the
increase in pathogen virulence and infection severity (Macarisin
et al., 2007).
2.2. Catalase production as a fungal infection strategy
Macarisin et al. (2007) also evaluated the effects of catalase
addition (CAT) on the infection incidence and lesion diameters.
These researchers reported that 500 U/mL catalase was sufficient to
promote a significant increase in the lesion diameter of samples
infected with P. digitatum and, to a lesser extent, with P. expansum
(Macarisin et al., 2007). Catalase is an antioxidant enzyme pro-
duced by fungi (Chaga et al., 1992; Aguirre et al., 2006; Ballester
et al., 2006), and it decomposes hydrogen peroxide to water and
oxygen. Therefore, it is suggested that the H2O2 that is produced by
the plant as a defense mechanism is decomposed by the catalase
secreted by P. digitatum (Macarisin et al., 2007; Ballester et al.,
2006) as an infection strategy.
2.3. Organic acid secretion by the pathogen for pH modulation
An interesting strategy used by the pathogen to infect the host is
the secretion of organic acids resulting in pH modulation. Con-
trolling the pH has been considered a factor and a regulatory cue
used by P. expansum, P. digitatum, and P. italicum for increased
pathogenicity. The authors verified a pH difference of 1.65 between
the healthy oranges (pH 4.77 ± 0.45) and decayed oranges (pH
3.12 ± 0.07) infected by P. digitatum. Prusky et al. (2004) suggested
that P. digitatum could enhance its pathogenicity through local pH
modulation of hosts, leading to an optimal pH for specific cell wall-
degrading enzymes, such as polygalacturonases (PG) (Barmore and
Brown, 1981; Prusky et al., 2004; Zhang et al., 2013).
2.4. Flavonoid production in citrus fruits as a natural defense
barrier
Another factor that influences the P. digitatum-citrus interaction
is the production of flavonoids by the host as defense barriers.
Flavonoids are well-documented as part of plant defense in Citrus
spp. as reported by different studies (Ortun ~ o and Del Río, 2009;
Ortun~ o et al., 2011; Kim et al., 2011). Flavonoids act as a chemical
barrier in the outermost citric tissue and can be considered phy-
toalexins against P. digitatum, such as the polymethoxyflavones
localized at the flavedo level of fruits and the flavanones localized at
the albedo level (Fig. 1) (Ortun ~ o and Del Río, 2009; Ortun~ o et al.,
2011). Studies indicate changes in the levels of flavanone glyco-
side - hesperidin (decreased approximately 7.8 %) and its corre-
sponding aglycon - hesperetin (increased approximately 7.2 %) at
five days after inoculation with P. digitatum. Similar results were
also found in lemons and grapefruits and can be associated with the
hydrolyzing action of P. digitatum (Ortun ~ o and Del Río, 2009;
Ortun~ o et al., 2011).
Ortun~ o et al. (2006) described polymethoxyflavones 5,6,7,30 ,40 -
pentamethoxyflavone (sinensetin), 5,6,7,8,30 ,40 -hexamethoxy-
flavone (nobiletin), 3,5,6,7,8,30 ,40 -heptamethoxyflavone and
5,6,7,8,40 -pentamethoxyflavone (tangeretin) as important com-
pounds in citrus defense against pathogens. Furthermore, these
researchers observed in vitro that nobiletin followed by hesperidin
and naringin were the flavonoids that exhibited higher activities
against P. digitatum growth on potato dextrose agar (PDA) medium
 19
J.H. Costa et al. / Fungal Biology 123 (2019) 584e593

Fig. 1. Natural product (NP) levels on citric fruits before and after interaction with Penicillium digitatum - NP shown in black are from citric fruits, which decrease the level in
approximately 7e8 % (hesperidin and naringin) before P. digitatum inoculation or increase the level in approximately 7 % (nobiletin) after P. digitatum inoculation. NP shown in gray
color are at a low level. NP shown in green color (hesperetin, naringenin and 6,7-dimethoxy coumarin) were detected only after inoculation of Penicillium digitatum in fruit, which
was in keeping with previously published findings (Ortun ~ o et al., 2011). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of
this article.)

compared to 5,6,7,30 ,40 -pentamethoxyflavone and tangeretin
(Ortun ~ o et al., 2011). Interestingly, the corresponding aglycones are
more active as fungistatic agents against P. digitatum than the gly-
cosylated flavanones (Ortun ~ o and Del Río, 2009), which is in
contrast to the hydrolyzing action of P. digitatum and requires
further investigation.
In addition, citrus defense mechanisms studies against the
phytopathogen P. digitatum demonstrated that scoparone, an
antifungal phenolic compound derived from phenylpropanoid
metabolism, plays an important role as a citric phytoalexin present
in oil glands (Kim et al., 1991; Ben-Yehoshua et al., 2008). Other
studies on the role of flavonoids in the defense mechanisms of
citrus fruits against the P. digitatum can be checked in the following
references (Del Río et al., 1998, 2004; Kim et al., 2011; Ballester
et al., 2013).
2.5. Metabolomic and transcriptomic analysis of citrus fruits
infected by P. digitatum
Tang et al. (2018) performed a search for the mechanism of
deterioration of Powell orange pulps from Citrus sinensis infected by
P. digitatum after 40 h and 60 h using GCeMS to determine the
levels of 26 polar primary metabolites. The authors also used the
headspaced solid phase microextraction and the gas-
chromatography-mass spectrometry (HS-SPME-GC-MS) technique
for the determination of 48 volatile organic compounds (VOCs). The
authors verified a change in the levels of these VOCs, similar to
octanoic acid ethyl ester, ethanol and aldehydes, which increased
the level in citrus fruits infected, suggesting that there is metabolic
reprogramming during P. digitatum infection. It was also observed
that jasmonates and ethylene pathways are involved in the plant
defense response for this interaction, as reported previously in
other pathogen-host interactions (Tang et al., 2018). In addition,
Tang et al. (2018) performed a transcriptomic analysis of
P. digitatum infection and verified that there was activation of some
transcription factors: 29 MYB, 25 WRKY and 33 AP2/ERF were
upregulated in fruits after 60 h of infection, suggesting their
involvement in the fruits' defense against the pathogen (Tang et al.,
2018).
2.6. Production of natural products by the fungi and pathogenicity
There is evidence that several phytopathogenic fungi have the
ability to produce small molecules able to decrease plant defense
responses, causing necrotic reactions in the infected cells
(Morrissey and Osbourn, 1999). For instance, the virulence of
several fungi (Cochliobolus heterostrophus, C. miyabeanus, Fusarium
graminearum and Alternaria brassicicola) on their respective host
plants has been reported to be mediated by particular siderophores,
a class of secondary metabolites involved in iron uptake (Fox and
Howlett, 2008).
In this context, during P. digitatum infection, Ariza et al. (2002)
identified secondary metabolites produced in P. digitatum
biomass: the indole alkaloids tryptoquialanine A (1) and trypto-
quialanine B (2) and the steroids cholesterol (3), ergosta-7,22-dien-
3b-OH (4), ergosta-7,22,24(28)-trien-3b-OH (5), episterol (6), and
eburicol (7), with (1) and (2) being reported as major secondary
metabolites in P. digitatum (Fig. 2). Secondary metabolites produced
by various microorganisms can contribute to the pathogenicity of
several pathogenic fungi (Scharf et al., 2014). However, trypto-
quialanine A (1), tryptoquialanine B (2) and other P. digitatum
secondary metabolites have not been directly related to the path-
ogenicity of P. digitatum (Zhu et al., 2017); therefore, the exact
biological role of these secondary metabolites has not been deter-
mined to date.
Attempting to investigate the biological role of tryptoquialanine
A (1), Costa et al. (2019) recently analyzed the surface of oranges
with green mold disease through mass spectrometry imaging and
observed that tryptoquialanines are the major metabolites pro-
duced during the decay process on the orange surface. Costa et al.
(2019) performed insecticidal bioassays using tryptoquialanine A
and observed high toxicity against Aedes aegypti larvae. Larvae
exposed to tryptoquialanine A presented a mortality rate of 37 %
within 24 h and of 81 % within 96 h, indicating an important
insecticidal action during the infection process. These insecticidal
assays suggest that the tryptoquialanines are involved in the pro-
tection of the pathogen and the rotten citrus against insects, which
explains the production of the tryptoquialanines by P. digitatum on
the citrus surface (Costa et al., 2019). Based on these results, it is
 20
J.H. Costa et al. / Fungal Biology 123 (2019) 584e593

Fig. 2. Chemical structure of tryptoquialanine A (1), tryptoquialanine B (2), cholesterol (3), ergosta-7,22-dien-3b-OH (4), ergosta-7,22,24(28)-trien-3b-OH (5), episterol (6), and
eburicol (7), tryptoquialanine C (8), tryptoquialanone (9), 15-dimethyl-2-epi-fumiquinazoline A (10), deoxytryptoquialanone (11), tryptoquivaline L (12), tryptoquivaline Q (13),
fumiquinazoline A (14) and fumiquinazoline C (15) isolated from Penicillium digitatum biomass.

possible to provide the first insight into the biological role of these
indole alkaloids involved in the P. digitatum-citrus interaction,
suggesting that tryptoquialanine A is an effective biocontrol agent
against insects during orange decay.
In addition, Costa et al. (2019) reported for the first time for
P. digitatum the production of a new tryptoquialanine, tryptoquia-
lanine C (8), and of intermediates involved in the tryptoquialanine
A biosynthetic pathway, such as tryptoquialanone (9), 15-dimethyl-
2-epi-fumiquinazoline A (10), and deoxytryptoquialanone (11).
Other secondary metabolites, such as tryptoquivaline L (12), tryp-
toquivaline Q (13), fumiquinazoline A (14) and fumiquinazoline C
(15), have also been reported (Costa et al., 2019).
The discovery of metabolites produced by P. digitatum and the
understanding of their biological role can provide more informa-
tion on the infection process and how this fungus may survive to
environmental pressures in the host. Once the P. digitatum genome
has been entirely sequenced, there is abundant information on the
great potential for secondary metabolite production. However,
notably few studies have been able to correlate these metabolites to
a biological role in the infection process.
During this process, the volatiles emitted from the ruptured oil
glands in wounded peel tissue can facilitate infection by promoting
spore germination and germ tube elongation of P. digitatum (Droby
et al., 2008).
Ariza et al. (2002) aimed to understand the citrus-pathogen
interaction by comparing the volatile profile of healthy oranges,
oranges submitted to mechanical damage and oranges infected
with P. digitatum. These researchers observed higher amounts of
limonene and other known citrus monoterpenes that were released
(instead of sesquiterpenes typical of healthy fruits) in mechanically
injured fruits than healthy ones. Other studies concerning the role
of limonene in the infection process and spore germination can be
checked in the following references (Rodríguez et al., 2015; Tao
et al., 2019). Regarding oranges contaminated with P. digitatum,

J.H. Costa et al. / Fungal Biology 123 (2019) 584e593
understand citrus colonization by P. digitatum since the whole
mechanism of infection has not been determined to date (Vu et al.,
2018; Zhu et al., 2017; Zhang et al., 2013b).
3.1. Transporters
The putative sucrose uptake transporter (SUT) gene PdSUT1 was
disrupted in P. digitatum (Ramo  n-Carbonell and Sa nchez-Torres,
2017b). Initially, the SUTs were considered specific to plants;
however, Schizosaccharomyces pombe was the first fungus found
with the homologous gene Sut1p (Ramo  n-Carbonell and Sa nchez-
Torres, 2017b; Reinders and Ward, 2001). SUTs have also been re-
ported in other fungi and related to virulence, but the main func-
tion and regulation of fungal sucrose uptake transporters have not
been thoroughly elucidated to date (Ramo n-Carbonell and
Sanchez-Torres, 2017b). DPdSUT1 mutants had lower virulence
than the wild type in citrus infection, while mutants with over-
expression in PdSUT1 did not present an increase in virulence
(Ramo n-Carbonell and Sa nchez-Torres, 2017b).
Expression of the genes involved in the major facilitator su-
perfamily (MFS) transporters was also analyzed by Ramo  n-
Carbonell and Sa nchez-Torres (2017b) and revealed that this
expression is enhanced in the DPdSUT1 mutants, which may
explain the increase in resistance to fungicides since the MFS car-
riers are powered and capable of carrying a wide variety of com-
pounds, such as toxic products and drugs (Ramo  n-Carbonell and
Sanchez-Torres, 2017b, Law et al., 2008; Reddy et al., 2012).
Ramo n-Carbonell and Sa nchez-Torres (2017b) concluded that the
PdSUT1 gene is not directly involved in virulence and resistance to
fungicides in P. digitatum. Instead, this gene only contributes to
these factors because it activates the expression of the MFS trans-
porter genes.
The ATP-binding cassette (ABC) and MFS transporters are
involved in virulence by mediating the secretion of host-specific
toxins or providing protection against plant defense components
(Stergiopoulos et al., 2002). ABC transporter genes in P. digitatum
are conserved and directly involved in drug resistance (Sa nchez-
Torrez and Tuset, 2011; Sun et al., 2013). In contrast, two MFS
transporters were studied and related to fungicide resistance,
although there are more than one hundred MFS genes in
P. digitatum (Sun et al., 2013). Disruption of the MFS genes PdMfs1
(Wang et al., 2012) and PdMfs2 (Wu et al., 2016) in P. digitatum leads
to mutants that are more sensitive to fungicides and less virulent in
citrus fruits.
The mechanism by which MFS plays a role during P. digitatum
infection requires further investigation (Wang et al., 2012). One of
the possibilities is that a toxin transported by MFS acts as a viru-
lence factor (Wang et al., 2012) or that MFS plays a role in infection
through a different way. For example, Liu et al. (2017) recently
described that the ChMfs1 gene of the phytopathogenic fungus
Colletotrichum higginsianum is involved in intrahyphal hyphae for-
mation, demonstrating a novel function of MFS transporters that
plays a key role in infection.
3.2. Cell wall-degrading enzymes
Phytopathogenic fungi need to pass through the plant cell wall,
an important barrier, to attack. Thus, phytopathogenic fungi pro-
duce cell wall-degrading enzymes (CWDEs) to depolymerize plant
cell wall structures, such as cellulose and pectin (Kubicek et al.,
2014). The ability to degrade the host cell wall has long been
thought to play a key role in P. digitatum infection such that CWDEs
are upregulated during the infection process (Zhang et al., 2013c).
Zhang et al. (2013c) verified the importance of the PdSNF1 gene
for virulence. In microorganisms, SNF1 regulates the derepression
22
of glucose-repressible genes, controlling the use of carbon sources
and, in the case of plant pathogenic fungi, the expression of CWDE
(Zhang et al., 2013c; Nadal et al., 2010; Yi et al., 2008). Concerning
virulence, Zhang et al. (2013c) observed that the DPdSFN1 mutants
were still pathogenic to citrus, but the symptoms took much longer
to develop in the fruits, and the production of conidia was affected.
Expression of the CWDE genes was upregulated for the wild-type
when induced with pectin, but in the DPdSNF1 mutants, the
expression levels almost did not change (Zhang et al., 2013c). Zhang
et al. (2013c) showed that the PdSNF1 gene is involved in the
regulation of CWDE gene expression in P. digitatum and, conse-
quently, in virulence.
CWDEs are also affected in P. digitatum by the mitogen-activated
protein kinase B gene PdMPkB. Through qRT-PCR analyses, Ma et al.
(2016) verified that CWDE genes were downregulated in DPdMPkB
mutants. Therefore, in fruits, DPdMPkB mutants showed fewer le-
sions than the wild type (Ma et al., 2016). The effect of pH signaling
transcription factor gene PdpacC deletion in P. digitatum was stud-
ied by Zhang et al. (2013a), and the disruption of this gene also
leads to fewer virulence mutants since the expression levels of the
polygalacturonase gene Pdpg2 and pectin lyase gene Pdpnl1 were
not upregulated or were weakly upregulated compared to the wild-
type strain. These results clearly reinforce that CWDEs are essential
for P. digitatum virulence as for other phytopathogens. Additionally,
genes that are involved in CWDE gene regulation are consequently
involved in pathogenicity.
3.3. Signaling pathway components
Studies on some signaling pathway genes demonstrated that
they are required for virulence but not directly involved in the
pathogenic mechanisms of P. digitatum. For example, PdpacC is a
signaling pathway gene that is essential for virulence because it
regulates CWDE gene expression (Zhang et al., 2013a).
Another example is the gene Pdos2, which encodes the mitogen-
activated protein kinase Hog1 studied by Wang et al. (2014) and is
part of the high-osmolarity glycerol pathway. DPdos2 mutants are
more sensitive to hyperosmotic stress and cell wall-disturbing
agents and are more resistant to fludioxonil once this fungicide
acts on the HOG pathway (Wang et al., 2014). Wang et al. (2014)
concluded that Pdos2 is associated with positive regulation of
glycerol synthesis and negative regulation of ergosterol synthesis,
and the decrease in virulence may result from the defect in vege-
tative growth and sensitivity to osmotic stress.
The role of the calcineurin-responsive transcription factor gene
PdCrz1 in P. digitatum was investigated by Zhang et al. (2013b).
DPdCrz1 mutants were observed to be defective in conidiation,
virulence, and hypersensitivity to stress caused by Ca2þ and DMI
fungicides. In addition, the expression of cell wall synthase genes
(CHS2, CHS3 and FKS1) leads to defective cell wall integrity (Zhang
et al., 2013). However, the mechanism by which PdCrz1 controls
P. digitatum virulence requires further investigation (Zhang et al.,
2013).
Asexual spores (conidia) are the inocula of green mold disease
produced by P. digitatum during fruit-pathogen interaction, and
they play an important role in the disease cycle since germination
of conidia occurs in the surface wounds of citrus (Zhang et al.,
2013c; Ramo  n-Carbonell and Sanchez-Torres, 2017a). The PdSNF1
gene, as mentioned before, is required for P. digitatum conidiation
(Zhang et al., 2013c). PdSNF1 plays a role in the expression of PdBrlA
and FadA, which are genes involved in conidiation and growth
signaling (Zhang et al., 2013c). PdMPkB and PdSlt2 are other genes
also involved in conidial formation (Ma et al., 2016; Ramo n-
Carbonell and Sa nchez-Torres, 2017c). Thus, DPdSNF1, DPdMPkB
and DPdSlt2 mutants showed less virulence compared to the wild-
 23
J.H. Costa et al. / Fungal Biology 123 (2019) 584e593

type strain (Zhang et al., 2013c; Ma et al., 2016; Ramo

 n-Carbonell
and Sanchez-Torres, 2017c) because conidiation was affected in
these mutants. The PdSte12 transcription factor is functionally
conserved in P. digitatum for asexual reproduction and is another
example of a signal pathway gene that contributes to the infection
process once it controls invasive growth and asexual reproduction
as the major virulence function (Ramo  n-Carbonell and Sanchez-
Torres, 2017a).
3.4. Structural components
Zhu et al. (2014) reported the function of glucosylceramides
(GlcCers) in P. digitatum through knockout of the glucosylceramide
synthase gene PdGcs1. For decades, GlcCers were considered only
structural components of the cell membrane; however, recent
studies describe that fungal glycosylceramides play important roles
in vital processes such as growth, spore germination, secretion, cell
wall assembly, and regulation of virulence (Mouyna et al., 2010;
Nimrichter et al., 2008). Deletion of PdGcs1 resulted in a decrease
in sporulation, growth and virulence and a delay in conidia
germination of P. digitatum (Zhu et al., 2014). Lesions caused by the
infection appear at 1 dpi for the wild-type strain and at 2 dpi for
DPdGcs1 mutants. At 5 dpi, the lesion diameter was approximately
11 cm for the wild-type strain and 5 cm for the deficient DPdGcs1
mutants (Zhu et al., 2014). For some fungi, the role of GlcCers in
virulence is well-established (Rittershaus et al., 2006); however, for
P. digitatum, this role has not been determined (Zhu et al., 2014).
Gandía et al. (2014) analyzed the effect of a Chitin synthase (Chs)
knockout in P. digitatum mutants. Chitin is a long linear
homopolymer made of N-acetylglucosamine bound by b-1,4 link-
age (Gandía et al., 2012, 2014, Ruiz-Herrera et al., 2002). This
polysaccharide is a structural component of the fungal cell walls,
giving rigidity to the cell and accounting for up to 20 % of the dry
weight of the cell wall of filamentous fungi (Gandía et al., 2012,
2014, Ruiz-Herrera et al., 2002). DPdChsVII mutants showed
reduced growth and conidia production and higher sensitivity to
cell wall-disturbing agents and fungicides (Gandía et al., 2014). The
virulence was reduced, but the mutants were still able to infect and
macerate the citrus, and there was no production of visible myce-
lium or green conidia in fruit (Gandía et al., 2014).
As a structural component, chitin synthases seem to be essential
for cell wall integrity, and the reduced virulence is a consequence.
The knockout of the O-mannosyltransferase protein gene Pdpmt2
by Harries et al. (2015) is another example. DPdpmt2 mutants were
more sensitive to cell wall breakers, showed a reduction in growth
and number of conidia and were less sensitive to PAF26 (Harries
et al., 2015). The slow growth and defective cell wall caused by
Pdpmt2 knockout limit fruit colonization and explain the lower
virulence of the mutants (Harries et al., 2015).
3.5. Genes not required for virulence
There are studies in which genes deleted in P. digitatum did not
alter virulence. It is the case of the PdMit1 gene encoding a
mannose inositol-phosphorylceramide synthase. The deletion of
PdMit1, evaluated by Zhu et al. (2015), altered growth, delayed
conidial germination, decreased conidial production and increased
Ca2þ sensitivity.

Table 1
Main genes studied and their role in P. digitatum metabolism.

Group Genes Role in P. digitatum References

Transporters
CWDE
Signal Pathways
Structural Components
Dispensable for virulence
PdSUT1 Related to virulence and fungicide sensitivity through MFS Ramo
 n-Carbonell and Sa
nchez-Torres (2017b)
transporters gene activation
PdMfs1 Required for prochloraz resistance, conidiation and full Wang et al. (2012)
virulence
PdMfs2 Involved in DMI resistance and pathogenicity Wu et al. (2016)
Pdpg2 Involved in virulence Zhang et al. (2013a)
PdpacC Participates in the response to Naþ and Kþ stresses, required Zhang et al. (2013a)
for mycelial growth at alkaline conditions. Involved in the
regulation of CWDEs genes Pdpg2 and Pdpnl1
Pdos2 Involved in response to hyperosmotic stress, regulation of Wang et al. (2014)
cell wall integrity, sensitivity to fungicides and virulence
PdCrz1 Involved in conidiation, virulence, DMI resistance and Zhang et al. (2013b)
expression of cell wall synthase genes
PdSNF1 Involved in virulence through the regulation of CWDEs, Zhang et al. (2013c)
PdBrlA and FadA genes expression.
PdMPkB Involved in the regulation of CWDEs genes expression, Ma et al. (2016)
conidia formation and osmotic stress adaptation.
PdSlt2 Involved in asexual reproduction and regulation of ABC and Ramo
 n-Carbonell and Sa
nchez-Torres (2017c)
MFS genes expression
PdSte12 Controls invasive growth and asexual reproduction Ramo  n-Carbonell and Sa
nchez-Torres (2017a);
Vilanova et al. (2016)
Pdac1 Required for vegetative growth, carbon utilization, and Wang et al. (2016)
conidial germination
PdChsVII Involved in cell wall integrity, vegetative growth and, Gandía et al. (2014)
during host colonization, mycelium development and
conidia production.
PdGcs1 Regulate cell growth, differentiation, and virulence by Zhu et al. (2014)
controlling the biosynthesis of GlcCers
Pdpmt2 Critical role in fungal growth, conidiation and PAF26 Harries et al. (2015)
resistance
PdsreA and PdsreB Important role in DMI resistance and global gene regulation Ruan et al. (2017)
PdMit1 Important for the growth of mycelium, sporulation and Zhu et al. (2015)
conidial germination.
tqaA Only involved in tryptoquialanines production Zhu et al. (2017)
PdKu80 Important role in homologous integration Xu et al. (2014)

J.H. Costa et al. / Fungal Biology 123 (2019) 584e593
Sterol regulatory element-binding protein genes (Pdsre) were
evaluated by Ruan et al. (2017) using three different P. digitatum
mutants. The lack mutants DPdsreA, DPdsreB and DPdsreAB caused
macerations of similar size to the lesions induced on mandarins by
the wild-type strain. Pdsre are principally involved in resistance to
DMI fungicides because DPdsre mutants increased sensitivity to
these fungicides (Ruan et al., 2017).
The mutants with deletions in tqaA, a gene involved in the
production of tryptoquialanines, showed no growth, development,
stress response or pathogenicity variations, as reported by Zhu et al.
(2017), indicating that tryptoquialanines are not involved in the
interaction of P. digitatum with citrus host. Zhu et al. (2017) sug-
gested that tryptoquialanines may be involved in the protection of
rotten citrus from insects. This hypothesis was confirmed, as
mentioned before by Costa et al. (2019), through the high insecti-
cide activity of tryptoquialanine A against A. aegypti larvae.
Similar results were also obtained for the deletion of PdKu80, a
gene of the nonhomologous end-joining (NHEJ) pathway (Xu et al.,
2014), and for ndo1, a gene coding a naphthalene dioxygenase
(Lo pez-Perez et al., 2015); none of these genes was found to be
important for virulence.
Table 1 summarizes the main genes studied for P. digitatum
found in the literature.
The functions of most of the studied P. digitatum genes are
conserved across the species. The genes that affect virulence are
generally related to growth, asexual reproduction, cell wall integ-
rity or the expression of other genes (CWDEs or transporters).
Some genes, such as PdCrz1, have not had their role in infection
thoroughly elucidated and warrant further investigation. As
mentioned previously, the functions of the transporters on viru-
lence have not been well-established, as no secondary metabolites
have been linked to the infection process, representing a gap in the
research concerning the fungus P. digitatum. The studies reinforced
that cell wall-degrading enzymes seem to be involved in the viru-
lence of this fungus and help to explain how P. digitatum modulates
the acidity of the environment. However, the molecular biology of
P. digitatum warrants a thorough investigation, and an extensive
analysis of the gene clusters found in the genome of this important
pathogen is necessary to better understand the entire infection
process.
4. Conclusions
Despite the great economic interest in P. digitatum, the molec-
ular basis of infection and specificity towards the citrus host re-
mains largely unknown. In recent years, most of the research on
P. digitatum has focused on treatments against infection symptoms,
fungicide resistance and the biocontrol of this fungus using
antagonist microorganisms. However, since 2012, with the publi-
cation of the genome of this important necrotrophic fungus, this
scenario has been changing, and many genes have been determined
to be virulence factors and to be important in the disease. The
knowledge concerning the genes required for full virulence
reviewed in this study (PdSNF1, PdMPkB, Pdos2, PdSte12, PdCrz1,
Pdpmt2, PdchsVII) provides the first insights about pathogen-host
interaction in P. digitatum and provides initial thoughts on poten-
tial target fungicides to be developed against such phytopathogens.
The main infection strategies of P. digitatum, citrus responses, sec-
ondary metabolites and genes involved in host-pathogen interac-
tion are represented in Fig. 3. In our opinion, understanding the
infection process and the defense responses of the hosts, as well as
the virulence mechanisms of this pathogen, is the first step in
developing safer and more eco-friendly alternative strategies for
controlling citrus postharvest diseases, leading to new and safer
control strategies for green mold disease. In addition, the secondary
24
metabolism of P. digitatum and the roles of natural products in the
infection merit further research. Innumerous biosynthetic gene
clusters are coded in this phytopathogen's genome, and many gene
clusters are described to be upregulated during the infection pro-
cess; however, only tryptoquialanine-like metabolites are
described as natural products for this fungus, meaning that many
products remain to be discovered.
Acknowledgments
This study was financed in part by the Coordenaç~ ao de Aper-
feiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance
Code 001 and Fundaça ~o de Amparo a Pesquisa no Estado de Sa ~o
Paulo [grant number FAPESP 2017/24462-4 and 2018/03670-0]. We
also thank Eder Vilhena Araujo for the orange photos used in Fig. 1.
References
Aguirre, J., Hansberg, W., Navarro, R., 2006. Fungal responses to reactive oxygen
species. Med. Mycol. 44, 101e107.
Apostol, I., Heinstein, P.F., Low, P.S., 1989. Rapid stimulation of an oxidative burst
during elicitation of cultured plant cells. Plant Physiol. 90, 109e116.
Ariza, M.R., Larsen, T.O., Petersen, B.O., Duus, J.Ø., Barrero, A.F., 2002. Penicillium
digitatum metabolites on synthetic media and citrus fruits. J. Agric. Food Chem.
50, 6361e6365.
Ballester, A.R., Lafuente, M.T., Gonza lez-Candelas, L., 2006. Spatial study of antiox-
idant enzymes, peroxidase and phenylalanine ammonia-lyase in the citrus
fruitePenicillium digitatum interaction. Postharvest Biol. Technol. 39, 115e124.
Ballester, A.R., Lafuente, M.T., Gonzalez-Candelas, L., 2013. Citrus phenylpropanoids
and defence against pathogens. Part II: gene expression and metabolite accu-
mulation in the response of fruits to Penicillium digitatum infection. Food Chem.
136, 285e291.
Barmore, C.R., Brown, G.E., 1981. Polygalacturonase from citrus fruit infected with
Penicillium italicum. Phytopathology 71, 328e331.
Ben-Yehoshua, S., Rodov, V., Nafussi, B., Feng, X., Yen, J., Koltai, T., Nelkenbaum, U.,
2008. Involvement of limonene hydroperoxides formed after oil gland injury in
the induction of defense response against Penicillium digitatum in lemon fruit.
J. Agric. Food Chem. 56, 1889e1895.
Brisson, L.F., Tenhaken, R., Lamb, C., 1994. Function of oxidative crosslinking of cell
wall structural proteins in plant disease resistance. Plant Cell 6, 1703e1712.
Chaga, G.S., Medin, A.S., Chaga, S.G., Porath, J.O., 1992. Isolation and characterization
of catalase from Penicillium chvysogenum. J. Chromatogr A. 604, 177e183.
Citrus: World Markets and Trade U.S. Department of Agriculture, Washington, DC.
https://apps.fas.usda.gov/psdonline/circulars/citrus.pdf (accessed 16 May 2019).
Conrath, U., 2006. Systemic acquired resistance. Plant Signal. Behav. 1, 179e184.
Costa, J.H., Bazioli, J.M., Araújo, E.V., Vendramini, P.H., Porto, M.C.F., Eberlin, M.N.,
Souza-Neto, J.A., Fill, T.P., 2019. Monitoring indole alkaloid production by
Penicillium digitatum during infection process in citrus by Mass Spectrometry
Imaging and molecular networking. Fungal Biol. 123 (8), 593e599.
Del Río, J.A., Arcas, M.C., Benavente-García, O., Ortun ~ o, A., 1998. Citrus poly-
methoxylated flavones can confer resistance against Phytophthora citrophthora,
Penicillium digitatum, and Geotrichum species. J. Agric. Food Chem. 46,
4423e4428.
Del Río, J.A., Go mez, P., Baidez, A.G., Arcas, M.C., Botía, J.M., Ortun ~ o, A., 2004.
Changes in the levels of polymethoxyflavones and flavanones as part of the
defense mechanism of Citrus sinensis (Cv. Valencia Late) fruits against Phy-
tophthora citrophthora. J. Agric. Food Chem. 52, 1913e1917.
Droby, S., Eick, A., Macarisin, D., Cohen, L., Rafael, G., Stange, R., McColum, G.,
Dudai, N., Nasser, A., Wisniewski, M., Shapira, R., 2008. Role of citrus volatiles in
host recognition, germination and growth of Penicillium digitatum and Penicil-
lium italicum. Postharvest Biol. Technol. 49 (3), 386e396.
Fox, E.M., Howlett, B.J., 2008. Secondary metabolism: regulation and role in fungal
biology. Curr. Opin. Microbiol. 11 (6), 481e487.
Frisvad, J.C., Samson, R.A., 2004. Polyphasic taxonomy of Penicillium subgenus
Penicillium A guide to identification of food and air-borne terverticillate Peni-
cillia and their mycotoxins. Stud. Mycol. 49, 1e174.
Gandía, M., Harries, E., Marcos, J.F., 2012. Identification and characterization of
chitin synthase genes in the postharvest citrus fruit pathogen Penicillium dig-
itatum. Fungal Biol. 116, 654e664.
Gandía, M., Harries, E., Marcos, J.F., 2014. The myosin motor domain-containing
chitin synthase PdChsVII is required for development, cell wall integrity and
virulence in the citrus postharvest pathogen Penicillium digitatum. Fungal
Genet. Biol. 67, 58e70.
García-Olmedo, F., Rodríguez-Palenzuela, P., Molina, A., Alamillo, J.M., Lo pez-
Solanilla, E., Berrocal-Lobo, M., 2001. Antibiotic activities of peptides, hydrogen
peroxide and peroxynitrite in plant defence. FEBS Lett. 498, 219e222.
Ghooshkhaneh, N.G., Golzarian, M.R., Mamarabadi, M., 2018. Detection and classi-
fication of citrus green mold caused by Penicillium digitatum using multispectral
imaging. J. Sci. Food Agric. 98 (9), 3542e3550.

J.H. Costa et al. / Fungal Biology 123 (2019) 584e593
Hao, W., Li, H., Hu, M., Yang, L., Rizwan-ul-Haq, M., 2011. Integrated control of citrus
green and blue mold and sour rot by Bacillus amyloliquefaciens in combination
with tea saponin. Postharvest Biol. Technol. 59, 316e323.
Harries, E., Gandía, M., Carmona, L., Marcos, J.F., 2015. The Penicillium digitatum
protein O-mannosyltransferase Pmt2 is required for cell wall integrity, con-
idiogenesis, virulence and sensitivity to the antifungal peptide PAF26. Mol.
Plant Pathol. 16, 748e761.
Hunt, M.D., Neuenschwander, U.H., Delaney, T.P., Weymann, K.B., Friedrich, L.B.,
Lawton, K.A., Steiner, H.Y., Ryals, J.A., 1996. Recent advances in systemic ac-
quired resistance research- a review. Gene 179, 89e95.
Ismail, M.A., Zhang, J., 2004. Post-harvest citrus diseases and their control. Outlooks
Pest Manag. 15, 29e35.
Julca, I., Droby, S., Sela, N., Marcet-Houben, M., Gabaldo, T., 2015. Contrasting
genomic diversity in two closely related postharvest pathogens: Penicillium
digitatum and Penicillium expansum. Genome. Biol. Evol. 8, 218e227.
Kanetis, L., Fo €rster, H., Adaskaveg, J.E., 2010. Determination of natural resistance
frequencies in Penicillium digitatum using a new air-sampling method and
characterization of Fludioxonil- and Pyrimethanil-Resistant isolates. Phytopa-
thology 100, 738e743.
Kim, J.J., Ben-Yehoshua, S., Shapiro, B., Henis, Y., Carmeli, S., 1991. Accumulation of
scoparone in heat-treated lemon fruit inoculated with Penicillium digitatum
Sacc. Plant Physiol. 97, 880e885.
Kim, H.G., Kim, G.S., Lee, J.H., Park, S., Jeong, W.Y., Kim, J.H., Kin, S.T., Cho, Y.A.,
Lee, W.S., Lee, S.J., Shin, S.C., 2011. Determination of the change of flavonoid
components as the defense materials of Citrus unshiu Marc. fruit peel against
Penicillium digitatum by liquid chromatography coupled with tandem mass
spectrometry. Food Chem. 128, 49e54.
Kubicek, C.P., Starr, T.L., Glass, N.L., 2014. Plant cell wall-degrading enzymes and
their secretion in plant-pathogenic fungi. Annu. Rev. Phytopathol. 52,
427e451.
Kuzniak, E., Urbanek, H., 2000. The involvement of hydrogen peroxide in plant
responses to stresses. Acta Physiol. Plant. 22, 195e203.
Ladanyia, M., 2010. Citrus Fruit: Biology, Technology and Evaluation, first ed. Aca-
demic Press. ISBN-10: 0123741300.
Lafuente, M.T., Alfe rez, F., Gonza lez-Candelas, L., 2018. Light-emitting diode blue
light alters the ability of Penicillium digitatum to infect citrus fruits. Photochem.
Photobiol. 94, 1003e1009.
Law, C.J., Maloney, P.C., Wang, D., 2008. Ins and outs of major facilitator superfamily
antiporters. Annu. Rev. Microbiol. 62, 289e305.
Levine, A., Tenhaken, R., Dixon, R., Lamb, C., 1994. H2O2 from the oxidative burst
orchestrates the plant hypersensitive disease resistance response. Cell 79,
583e593.
Liu, L., Yan, Y., Huang, J., Hsiang, T., Wei, Y., Li, Y., Gao, J., Zheng, L., 2017. A novel MFS
transporter gene ChMfs1 is important for hyphal morphology, conidiation, and
pathogenicity in Colletotrichum higginsianum. Front. Microbiol. 8, 1953.
Lopez-Pe rez, M., Ballester, A.R., Gonz alez-Candelas, L., 2015. Identification and
functional analysis of Penicillium digitatum genes putatively involved in viru-
lence towards citrus fruit. Mol. Plant Pathol. 16, 262e275.
Louw, J.P., Korsten, L., 2019. Impact of ripeness on the infection and colonisation of
Penicillium digitatum and P. expansum on plum. Postharvest Biol. Technol. 149,
148e158.
Lu, H., Higgins, V.J., 1999. The effect of hydrogen peroxide on the viability of tomato
cells and of the fungal pathogen Cladosporium fulvum. Physiol. Mol. Plant
Pathol. 54, 131e143.
Ma, H., Sun, X., Wang, M., Gai, Y., Chung, K.R., Li, H., 2016. The citrus postharvest
pathogen Penicillium digitatum depends on the PdMpkB kinase for develop-
mental and virulence functions. Int. J. Food Microbiol. 236, 167e176.
Macarisin, D., Cohen, L., Eick, A., Rafael, G., Belausov, E., Wisniewski, M., Droby, S.,
2007. Penicillium digitatum suppresses production of hydrogen peroxide in host
tissue during infection of citrus fruit. Phytopathology 97, 1491e1500.
Marcet-Houben, M., Ballester, A.R., la Fuente, B., Harries, E., Marcos, J.F., Gonz alez-
Candelas, L., Galbado  n, T., 2012. Genome sequence of the necrotrophic fungus
Penicillium digitatum, the main postharvest pathogen of citrus. BMC Genomics
13, 646.
Mellersh, D.G., Foulds, I.V., Higgins, V.J., Heath, M.C., 2002. H2O2 plays different roles
in determining penetration failure in three diverse plant-fungal interactions.
Plant J. 29, 257e268.
Molto  , E., Blasco, J., Go
mez-Sanchıs, J., 2010. Analysis of hyperspectral images of
citrus fruits. In: Sun, D.W. (Ed.), Hyperspectral Imaging for Food Quality Anal-
ysis and Control. Academic Press, pp. 321e348.
Morrissey, J.P., Osbourn, A.E., 1999. Fungal resistance to plant antibiotics as a
mechanism of pathogenesis. Microbiol. Mol. Biol. Rev. 63 (3), 708e724.
Mouyna, I., Kniemeyer, O., Jank, T., Loussert, C., Mellado, E., Aimanianda, V.,
Beauvais, A., Wartenberg, D., Sarfati, J., Bayry, J., Pre vost, M.C., Brakhage, A.A.,
Strahl, S., Huerre, M., Latge , J.P., 2010. Members of protein O-mannosyl-
transferase family in Aspergillus fumigatus differentially affect growth,
morphogenesis and viability. Mol. Microbiol. 76, 1205e1221.
Nadal, M., Garcia-Pedrajas, M.D., Gold, S.E., 2010. The snf1 Gene of ustilago maydis
acts as a dual regulator of cell wall degrading enzymes. Phytopathology 100,
1364e1372.
Nathues, E., Joshi, S., Tenberge, K.B., von den Driesch, M., Oeser, B., B€ aumer, N.,
Mihlan, M., Tudzynski, P., 2004. CPTF1, a CREB-like transcription factor, is
involved in the oxidative stress response in the phytopathogen Claviceps pur-
purea and modulates ROS level in its host Secale cereal. Phytopathology 17,
383e393.
25
Neuencchwander, U., Vernooij, B., Friedrich, L., Uknes, S., Kessmann, H., Ryals, J.,
1995. Is hydrogen peroxide a second messenger of salicylic acid in systemic
acquired resistance? Plant J. 8, 227e233.
Nimrichter, L., Rodrigues, M.L., Barreto-Bergter, E., Travassos, L.R., 2008. Sophisti-
cated functions for a simple molecule: the role of glucosylceramides in fungal
cells. Lipid Insights 2, 61e73.
Olson, P.D., Varner, J.E., 1993. Hydrogen peroxide and lignification. Plant J. 4,
887e892.
Ortun~ o, A., Del Río, J.A., 2009. Role of citrus phenolic compounds in the resistance
mechanism against pathogenic fungi. Tree For. Sci. Biotech. 3, 49e53.
Ortun~ o, A., B
aidez, A., Go  mez, P., Arcas, M.C., Porras, I., García-Lido n, A., Del Río, J.A.,
2006. Citrus paradisi and Citrus sinensis flavonoids: their influence in the
defence mechanism against Penicillium digitatum. Food Chem. 98, 351e358.
Ortun~ o, A., Díaz, L., Alvarez, N., Porras, I., García-Lido n, A., Del Río, J.A., 2011.
Comparative study of flavonoid and scoparone accumulation in different Citrus
species and their susceptibility to Penicillium digitatum. Food Chem. 125, 232e239.
Parveen, S., Wani, A.H., Bhat, M.Y., Koka, J.A., 2016. Biological control of postharvest
fungal rots of rosaceous fruits using microbial antagonists and plant extracts e
a review. Czech Mycol. 68, 41e46.
Perez, M.F., Ibarreche, J.P., Isas, A.S., Sepulveda, M., Ramallo, J., Dib, J.R., 2017.
Antagonistic yeasts for the biological control of Penicillium digitatum on lemons
stored under export conditions. Biol. Control 115, 135e140.
Poppe, L., Vanhoutte, S., Ho €fte, M., 2003. Modes of action of Pantoea agglomerans
CPA-2, an antagonist of postharvest pathogens on fruits. Eur. J. Plant Pathol. 109,
963e973.
Prusky, D., McEvoy, J.L., Saftner, R., Conway, W.S., Jones, R., 2004. Relationship be-
tween host acidification and virulence of Penicillium spp. on apple and citrus
fruit. Phytopathology 94, 44e51.
Qin, G., Liu, J., Cao, B., Li, B., Tian, S., 2011. Hydrogen peroxide acts on sensitive
mitochondrial proteins to induce death of a fungal pathogen revealed by pro-
teomic analysis. PLoS One 6, e21945.
Ramo n-Carbonell, M., Sa nchez-Torres, P., 2017a. The transcription factor PdSte12
contributes to Penicillium digitatum virulence during citrus fruit infection.
Postharvest Biol. Technol. 125, 129e139.
Ramo n-Carbonell, M., Sa nchez-Torres, P., 2017b. Involvement of Penicillium dig-
itatum PdSUT1 in fungicide sensitivity and virulence during citrus fruit infec-
tion. Microbiol. Res. 203, 57e67.
Ramo n-Carbonell, M., Sa nchez-Torres, P., 2017c. PdSlt2 Penicillium digitatum
mitogen activated-protein kinase controls sporulation and virulence during
citrus fruit infection. Fungal Biol. 121, 1063e1074.
Reddy, V.S., Shlykov, M.A., Castillo, R., Sun, E.I., Saier, M.H., 2012. The Major Facili-
tator Superfamily (MFS) Revisited,. FEBS J. 279 (11), 2022e2035.
Reinders, A., Ward, J.M., 2001. Functional characterization of the a-glucoside
transporter Sut1p from Schizosaccharomyces pombe, the first fungal homologue
of plant sucrose transporters. Mol. Microbiol. 39, 445e454.
Rittershaus, P.C., Kechichian, T.B., Allegood, J.C., Merrill, A.H., Hennig, M., Luberto, C.,
Del Poeta, M., 2006. Glucosylceramide synthase is an essential regulator of
pathogenicity of Cryptococcus neoformans. J. Clin. Investig. 116, 1651e1659.
Rodríguez, A., Shimada, T., Cervera, M., Redondo, A., Alque zar, B., Rodrigo, M.J.,
Zacarías, L., Palou, L., Lo pez, M.M., Pen ~ a, L., 2015. Resistance to pathogens in
terpene down-regulated orange fruits inversely correlates with the accumula-
tion of D-limonene in peel oil glands. Plant Signal. Behav. 10 (6) e1028704-
1ee1028704-4.
Ruan, R., Wang, M., Liu, X., Sun, X., Chung, K.R., Li, H., 2017. Functional analysis of
two sterol regulatory element binding proteins in Penicillium digitatum. PLoS
One 12 (5), e0176485.
Ruiz-Herrera, J., Gonza lez-Prieto, J.M., Ruiz-Medrano, R., 2002. Evolution and
phylogenetic relationships of chitin synthases from yeasts and fungi. FEMS
Yeast Res. 1, 247e256.
Sa
nchez-Torrez, P., Tuset, J.J., 2011. Molecular insights into fungicide resistance in
sensitive and resistant Penicillium digitatum strains infecting citrus. Postharvest
Biol. Technol. 59, 159e165.
Scharf, D.H., Heinekamp, T., Brakhage, A.A., 2014. Human and plant fungal patho-
gens: the role of secondary metabolites. PLoS Pathog. 10 (1), e1003859.
Stergiopoulos, I., Zwiers, L., De Waard, M.A., 2002. Secretion of natural and synthetic
toxic compounds from filamentous fungi by membrane transporters of the ATP-
binding cassette and major facilitator superfamily. Eur. J. Plant Pathol. 108 (7),
719e734.
Sun, X., Ruan, R., Lin, L., Zhu, C., Zhang, T., Wang, M., Li, H., Yu, D., 2013. Genomewide
investigation into DNA elements and ABC transporters involved in imazalil
resistance in Penicillium digitatum. FEMS Microbiol. Lett. 348, 11e18.
Talibi, I., Boubaker, H., Boudyach, E.H., Aoumar, A.A.B., 2014. Alternative methods for
the control of postharvest citrus diseases. J. Appl. Microbiol. 117, 1e17.
Tang, N., Chen, N., Hu, N., Deng, W., Chen, Z., Li, Z., 2018. Comparative metabolomics
and transcriptomic profiling reveal the mechanism of fruit quality deterioration
and the resistance of citrus fruit against Penicillium digitatum. Postharvest Biol.
Technol. 145, 61e73.
Tao, N., Chen, Y., Wu, Y., Wang, X., Li, L., Zhu, A., 2019. The terpene limonene induced
the green mold of citrus fruit through regulation of reactive oxygen species
(ROS) homeostasis in Penicillium digitatum spores. Food Chem. 277, 414e422.
Vilanova, L., Teixido  , N., Torres, R., Usall, J., Vin
~ as, I., Sa
nchez-Torres, P., 2016. Rele-
vance of the transcription factor PdSte12 in Penicillium digitatum conidiation
and virulence during citrus fruit infection. Int. J. Food Microbiol. 235, 93e102.
Vu, T.X., Ngo, T.T., Mai, L.T.D., Bui, T.-T., Le, D.H., Bui, H.T.V., Nguyen, H.Q., Ngo, B.X.,
Tran, V.-T., 2018. A highly efficient Agrobacterium tumefaciens-mediated

J.H. Costa et al. / Fungal Biology 123 (2019) 584e593
transformation system for the postharvest pathogen Penicillium digitatum using
DsRed and GFP to visualize citrus host colonization. J. Microbiol. Methods 144,
134e144.
Wang, J.Y., Li, H.Y., 2008. Agrobacterium tumefaciens-mediated genetic trans-
formation of the phytopathogenic fungus Penicillium digitatum. J. Zhejiang Univ.
Sci. B 9, 823e828.
Wang, J., Sun, X., Lin, L., Zhang, T., Ma, Z., Li, H., 2012. PdMfs1, a major facilitator
superfamily transporter from Penicillium digitatum, is partially involved in
the imazalil-resistance and pathogenicity. Afr. J. Microbiol. Res. 6 (1),
95e105.
Wang, M., Chen, C., Zhu, C., Sun, X., Ruan, R., Li, H., 2014. Os2 MAP kinase-mediated
osmostress tolerance in Penicillium digitatum is associated with its positive
regulation on glycerol synthesis and negative regulation on ergosterol syn-
thesis. Microbiol. Res. 169, 511e521.
Wang, W., Wang, M., Wang, J., Zhu, C., Chung, K., Li, H., 2016. Adenylyl cyclase is
required for cAMP production, growth, conidial germination, and virulence in the
citrus green mold pathogen Penicillium digitatum. Microbiol. Res. 192, 11e20.
Wani, S.H., 2010. Inducing fungus-resistance into plants through biotechnology.
Not. Sci. Biol. 2, 14e21.
Wu, Z., Wang, S., Yuan, Y., Zhang, T., Liu, J., Liu, D., 2016. A novel facilitator super-
family transporter in Penicillium digitatum (PdMFS2) is required for prochloraz
resistance, conidiation and full virulence. Biotechnol. Lett. 38 (1), 1349e1357.
Xu, Q., Zhu, C.Y., Wang, M.S., Sun, X.P., Li, H.Y., 2014. Improvement of a gene tar-
geting system for genetic manipulation in Penicillium digitatum. J. Zhejiang
Univ. Sci. B 15, 116e124.
26
Yi, M., Park, J.H., Ahn, J.H., Lee, Y.H., 2008. MoSNF1 regulates sporulation and
pathogenicity in the rice blast fungus Magnaporthe oryzae. Fungal Genet. Biol.
45, 1172e1181.
Zhang, T., Sun, X., Xu, Q., Candelas, L.G., Li, H., 2013a. The pH signaling transcription
factor PacC is required for full virulence in Penicillium digitatum. Appl. Micro-
biol. Biotechnol. 97, 9087e9098.
Zhang, T., Xu, Q., Sun, X., Li, H., 2013b. The calcineurin-responsive transcription
factor Crz1 is required for conidation, full virulence and DMI resistance in
Penicillium digitatum. Microbiol. Res. 168, 211e222.
Zhang, T., Sun, X., Xu, Q., Zhu, C., Li, Q., Li, H., 2013c. PdSNF1, a sucrose non-
fermenting protein kinase gene, is required for Penicillium digitatum con-
idiation and virulence. Appl. Microbiol. Biotechnol. 97, 5433e5445.
Zhu, C., Wang, M., Wang, W., Ruan, R., Ma, H., Mao, C., 2014. Glucosylceramides are
required for mycelial growth and full virulence in Penicillium digitatum. Bio-
chem. Biophys. Res. Commun. 455, 165e171.
Zhu, C., Wang, W., Wang, M., Ruan, R., Sun, X., He, M., Mao, C., Li, H., 2015. Deletion
of PdMit1, a homolog of yeast Csg1, affects growth and Ca2þ sensitivity of the
fungus Penicillium digitatum, but does not alter virulence. Res. Microbiol. 166,
143e152.
Zhu, C., Sheng, D., Wu, X., Wang, M., Hu, X., Li, H., Yu, D., 2017. Identification of
secondary metabolite biosynthetic gene clusters associated with the infection
of citrus fruit by Penicillium digitatum. Postharvest Biol. Technol. 134, 17e21.
 27

2 OBJETIVOS
Considerando a importância da citricultura para o Brasil e os prejuízos
causados pela doença do bolor verde, o objetivo desta tese é avançar no
entendimento dos mecanismos de patogenicidade do fungo P. digitatum. Para tal fim,
ferramentas e técnicas comumente utilizadas na área de produtos naturais (redes
moleculares, IMS e co-cultivos) serão empregues no sistema P. digitatum-hospedeiro.
Serão feito estudos e bioensaios visando a melhor compreensão do metabolismo e
do papel biológico dos metabólitos secundários de P. digitatum. Ao final desta tese,
espera-se contribuir com o desenvolvimento de estratégias mais seguras para o
controle deste fitopatógeno.
 28

3 CAPÍTULO I
Monitoramento da produção de alcalóides indólicos por Penicillium digitatum
durante o processo de infecção em citros utilizando imagem de espectrometria
de massas e rede molecular

O conteúdo deste capítulo é composto pelo artigo intitulado “Monitoring indole

alkaloid production by Penicillium digitatum during infection process in
citrus by Mass Spectometry Imaging and molecular networking”, publicado
no periódico Fungal Biology. A reprodução deste documento pelos autores, para
fins não comerciais, é autorizada pela editora Elsevier e não necessita de
permissão escrita (anexo 10.1).

Referência: Costa, J.H., Bazioli, J.M., Araújo, E.V., et al. 2019. Monitoring indole
alkaloid production by Penicillium digitatum during infection process in citrus by
Mass Spectometry Imaging and molecular networking. Fungal Biology, v. 123, p.
594-600. doi: 10.1016/j.funbio.2019.03.002

Versão final publicada disponível em:

https://www.sciencedirect.com/science/article/pii/S1878614619300352

3.1 Resumo
Neste trabalho a técnica de IMS foi aplicada para monitorar a produção de
metabólitos secundários produzidos na superfície de laranjas durante a infecção por
P. digitatum. O objetivo era obter informações sobre o mecanismo de interação entre
fungo-hospedeiro. Através da combinação entre IMS e redes moleculares foi possível
reportar, pela primeira vez, a produção de triptoquivalinas e fumiquinazolinas por P.
digitatum e a acumulação das triptoquialaninas na superfície dos frutos de 4 a 7 dias
após a infecção. O uso das redes moleculares em combinação com o banco de dados
de GNPS auxiliou na identificação de novos metabólitos produzidos por P. digitatum
in vivo e permitiu a comparação do perfil metabólico fúngico em diferentes
hospedeiros cítricos. O papel biológico das triptoquialaninas foi investigado. A
triptoquialanina A foi submetida a bioensaios inseticidas que demonstraram sua alta
toxicidade contra larvas de Aedes aegypt, sugerindo uma importante ação inseticida
durante o apodrecimento dos frutos.
 29

3.2 Artigo: “Monitoring indole alkaloid production by Penicillium digitatum during infection process in citrus by
Mass Spectrometry Imaging and molecular networking”
Contents lists available at ScienceDirect

Fungal Biology
journal homepage: www.elsevier.com/locate/funbio

Monitoring indole alkaloid production by Penicillium digitatum during

infection process in citrus by Mass Spectrometry Imaging and
molecular networking
Jonas Henrique Costa a, Jaqueline Moraes Bazioli a, b, Eder de Vilhena Araújo a,
Pedro Henrique Vendramini a, Mariana Cristina de Freitas Porto c,
Marcos Nogueira Eberlin a, Jayme A. Souza-Neto c, Taícia Pacheco Fill a, *
a
Institute of Chemistry, Universidade Estadual de Campinas, CP 6154, 13083-970, Campinas, SP, Brazil
b
Faculty of Pharmaceutical Sciences, Universidade Estadual de Campinas, 13083-859 Campinas, SP, Brazil
c ~
Sao Paulo State University (UNESP), School of Agricultural Sciences, Department of Bioprocesses and Biotechnology, Central Multiuser Laboratory,
Botucatu, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: Green mold, caused by Penicillium digitatum, is the most destructive post-harvest disease in citrus.
Received 15 January 2019 Secondary metabolites produced by fungal phytopathogens have been associated with toxicity to their
Received in revised form respective host through the interaction with a wide range of cell targets. Natural products have also been
26 February 2019
described as important molecules for biocontrol and competition in their respective environment. For
Accepted 6 March 2019
Available online 22 March 2019
P. digitatum, the production of indole alkaloids, tryptoquialanines A and B, have been reported. However,
their biological role remains unknown. Mass Spectrometry Imaging (MSI) technique was applied here for
Corresponding Editor: Gustavo Henrique the first time to monitor the secondary metabolites produced on the orange surface during infection in
Goldman order to gain insights about the P. digitatum-citrus interaction mechanisms. Through the combination of
MSI and molecular networking it was possible to report, for the first time, the production of trypto-
Keywords: quivalines and fumiquinazolines by P. digitatum and also the accumulation of tryptoquialanines on the
Alkaloids fruit surface from 4 to 7 d post inoculation. P. digitatum was also evaluated concerning the ability to
Citrus green mold sinthesize indole alkaloids in vivo in the different citrus hosts. The biological role of tryptoquialanines
Insecticidal activity
was investigated and tryptoquialanine A was submitted to insecticidal bioassays that revealed its high
Secondary metabolites
toxicity against Aedes Aegypti, suggesting an important insecticidal action during orange decay.
© 2019 British Mycological Society. Published by Elsevier Ltd. All rights reserved.

1. Introduction the phytopathogen (Macarisin et al., 2007), resulting in pH control,

Penicillium digitatum is the most severe postharvest pathogen of
citrus fruits and has been reported to cause the green mold disease,
contributing to up to 90 % of the total post harvest losses in tropical
sub-climates (Ghooshkhaneh et al., 2018; Macarisin et al., 2007).
P. digitatum is able to infect citrus fruits through pores and wounds
caused by environmental factors, or during harvest process and
transport (Perez et al., 2017).
There are several factors that mediate the hostepathogen
interaction between P. digitatum and citrus fruit, including
hydrogen peroxide modulation and secretion of organic acids by
* Corresponding author.
E-mail address: taicia@unicamp.br (T.P. Fill).
leading to an optimal pH for specific cell wall degrading enzymes,
such as polygalacturonases (PG) (Barmore and Brown, 1981; Prusky
et al., 2004; Zhang et al., 2013). Nevertheless, secondary metabo-
lites directly associated with the infection process in situ have not
been identified so far (Zhu et al., 2017).
Ariza et al. (2002) reported indole alkaloids, such as trypto-
quialanines A and B, as secondary metabolites of P. digitatum, which
are biosynthesized on citrus fruits during infection. The authors
also reported that the fungal infection on natural substrates
induced the production of citrus monoterpenes together with
fungal volatiles, such as sesquiterpenes. However, molecular
studies conduced by Zhu et al. (2017) indicated that tryptoquiala-
nines are not related to P. digitatum's pathogenicity in citrus fruits.
The deletion of tqaA gene (nonribossomal peptide synthetase) did
not affect the virulence of the gene knockout mutant when

https://doi.org/10.1016/j.funbio.2019.03.002
1878-6146/© 2019 British Mycological Society. Published by Elsevier Ltd. All rights reserved.

J.H. Costa et al. / Fungal Biology 123 (2019) 594e600
compared to the wild type strain, so the exact biological role of
tryptoquialanines is still unknown (Zhu et al., 2017).
The studies concerning this pathogen-host system mainly focus
on the fruit response to the pathogen or on the fungicide resis-
tance (Lope z-Perez et al., 2015), so the pathogenicity mechanisms
and infection process are still unclear (Zhu et al., 2017). Therefore,
it become necessary to seek a better understanding of these
mechanisms in different citrus fruits, in order to discover sec-
ondary metabolites potentially associated with the citrus infection.
Understanding the infection process and the fungus strategies is
an important step to develop ways to protect citrus from
P. digitatum infections leading to a more productive citriculture
worldwide.
In this context, Mass Spectrometry Imaging (MSI) is a powerful
and useful tool to describe the functional roles of secondary me-
tabolites in a biological context, based on their relative abundances
of m/z and spatial distribution (Lei et al., 2011). Through MSI it is
possible to investigate and identify secondary metabolites involved
at different stages of a phytopathogen infection. Here we explored
different P. digitatum infection stages by Mass Spectrometry Im-
aging combined with Molecular Networking using GNPS approach,
to profile the secondary metabolite production of P. digitatum
during infection process. The biological role of indole alkaloids
during this process is also discussed.
2. Materials and methods
2.1. Fungal culture
The P. digitatum strain used in the studies is deposited with the
Spanish Type Culture Collection (CECT) under the accession code
CECT20796. P. digitatum was maintained on commercial potato
dextrose agar (PDA) (Acumedia). PDA was autoclaved at 103 KPa
(121  C) for 15 min. PDA plates were stored at 25  C for 7 d in
darkness. Spores were harvested by washing the agar surface with
sterile distilled water and subsequently diluted to a final concen-
tration of 106 spore mL 1.
2.2. Mass spectrometry imaging (MSI)
MSI analysis were performed directly on the orange (Citrus
sinensis) peel surface infected with P. digitatum using a Prosolia
DESI source Modelo Omni Spray 2D®-3201) coupled to a Thermo
Scientific QExactive® Hybrid Quadrupole-Orbitrap Mass Spec-
trometer. The DESI configuration used was the same set by
Angolini et al. (2015) with small modifications. The methanol
flow rate was set at (10.0 mL min 1). MS data was processed with
Xcalibur software (version 3.0.63) developed by Thermo Fisher
Scientific. The DESI-MSI data was converted into image files
using Firefly data conversion software (version 2.1.05) and
viewed using BioMap software (version 3.8.0.4) developed by
Novartis Institutes for BioMedical Research. In BioMap, color
scaling was adjusted to a fixed value for comparison between
the samples.
2.3. Large scale experiment for isolation of compounds 1, 3 and 7
Large scale cultivation was performed by cultivating P. digitatum
in 12 L of PD (Potato-Dextrose) liquid media in sterile 1 L Erlen-
meyer and stored at 25  C for 10 d in darkness. After incubation,
culture broth was extracted twice with equal volumes of ethyl ac-
etate 100 % (1:1 v/v) and vacuum filtered. Solvent was removed
under reduced pressure and the final extract stored at 20  C.
30
2.4. Isolation of secondary metabolites 1, 3 and 7
High performance liquid chromatography (HPLC) separations
for 1, 3 and 7 were performed on a Phenomenex column Luna 5 mm
Phenyl-Hexyl (250  4.6 mm) using a SHIMADZU prominence HPLC
LC-20AT, equipped with CBM-20A communication bus module,
SPD-M20A photodiode array detector and SIL-20A auto sampler.
The mobile phase was water (A) and acetonitrile with 0.1 % (v/v) of
formic acid (B). Flow rate was 1.0 mL min 1. The eluent profile (A:B)
was: 0e50 min, gradient from 65:35 to 50:50; 50e70 min, gradient
from 50:50 to 40:60. Preparative HPLC purifications for 1, 3 and 7
were performed on a Phenomenex column Luna 5 mm Phenyl-Hexyl
(250  10 mm) using a Waters 1525 Binary HPLC Pump equipped
with Waters 2998 Photodiode Array Detector and Waters Fraction
Collector III using the same optimized gradient conditions with a
flow rate set at 4.7 mL min 1.
2.5. Characterization of secondary metabolites 1, 3 and 7
1
H NMR, 13C NMR and 2D experiments were acquired in a
Bruker Avance III 500 (1H 500.13 MHz and 13C 125.7 MHz).
Deuterated chloroform (CDCl3; 7.23 ppm), dimethyl dulfoxide
(DMSO; 2.50 ppm and 39.51 ppm) and tetramethylsilane (TMS;
0.0 ppm) were used as a solvent and internal reference. Chemical
shifts (d) were expressed in (ppm) and the coupling constants (J) in
Hertz (Hz).
2.6. In vivo and in vitro assays
For in vivo assays, mature oranges (Citrus sinensis), sicilian
lemons (Citrus limon) and tangerine (Citrus reticulata) obtained
from a local grocery store (Campinas, SP, Brazil) were surface-
sterilized with 2 % (v/v) sodium hypochlorite solution
(Panebianco et al., 2014), rinsed with distilled water and air-dried at
room temperature. The fruits were wounded at the equatorial re-
gion (1 cm wide x 1 cm deep) and 15 mL of P. digitatum spore so-
lution was inoculated in the wound site. Infected and control fruits
were stored in sterile 500 mL beakers. For in vitro assays, 15 mL of
spore solution were inoculated in 100 mL of sterile orange juice or
in 20 mL of PDA. All assays were done in duplicate and stored at
25  C for 7 d in darkness. After incubation time, the fruits peel were
cut (2 cm  2 cm) and the extractions of the secondary metabolites
produced in the fruits were made with 5 mL ethyl acetate 100 %
during 1 h in ultrasonic bath. Orange juice and PDA were extracted
twice with equal volumes of ethyl acetate 100 % (1:1 v/v) during 1 h
in ultrasonic bath. All the extracts were filtered, dried under N2 and
stored at 20  C.
2.7. MS/MS analysis
Crude extracts were resuspended in MeOH HPLC and centri-
fuged at 13,000 rpm for 5 min. The samples were analyzed using a
LC Agilent 1200 mass spectrometer coupled with Agilent iFunnel
6550 Q-ToF LC-MS. The electrospray ionization source operated in
positive mode ESI (þ), following operating conditions: nebulizing
gas temperature: 290  C; capillary voltage: þ3500 V; nozzle
voltage: 320 V; drying gas flow: 12 mL min 1; nebulization gas
pressure: 50 psi; auxiliary gas temperature: 350  C and flow of
auxiliary gas: 12 mL min 1. The analyzer time of flight (ToF) oper-
ated in the range m/z 50e1500. Collision Energy formula (auto MS/
MS mode): 4 V (slope)*(m/z)/100 þ 5 V (offset). Maximum 5 pre-
cursors per cycle were selected. 2 mL of sample were injected. Sta-
tionary phase: Thermo Scientific column Accucore C18 2.6 mm,
2.1 mm  100 mm. The mobile phase was water (A) and acetonitrile
with 0.1 % (v/v) of formic acid (B). Flow rate was 0.2 mL min 1. The

J.H. Costa et al. / Fungal Biology 123 (2019) 594e600
eluent profile (A:B) was: 0e10 min, gradient from 95:5 to 2:98;
10e15 min, isocratic elution with 2:98; 15e16.2 min, gradient from
2:98 to 95:5; 16.2e20 min, isocratic elution with 95:5. The spectra
were processed with Agilent Mass Hunter Workstation Software.
2.8. Molecular MS/MS network
A molecular network for P. digitatum cultivated in sicilian lemon,
tangerine, orange, orange juice and PDA was created using the
online workflow at GNPS (http://gnps.ucsd.edu). The data was
filtered by removing all MS/MS peaks within ±17 Da of the pre-
cursor m/z. MS/MS spectra were window filtered by choosing only
the top 6 peaks in the ± 50 Da window throughout the spectrum.
The data was then clustered with MS-Cluster with a parent mass
tolerance of 0.2 Da and a MS/MS fragment ion tolerance of 0.1 Da to
create consensus spectra. Further, consensus spectra that contained
less than 2 spectra were discarded. A network was then created
where edges were filtered to have a cosine score above 0.7 and
more than 6 matched peaks. Further edges between two nodes
were kept in the network only if each of the nodes appeared in each
other's respective top 10 most similar nodes. The spectra in the
network were then searched against GNPS0 spectral libraries. The
library spectra were filtered in the same manner as the input data.
All matches kept between network spectra and library spectra were
required to have a score above 0.7 and at least 6 matched peaks.
2.9. Evaluation of insecticidal activity of tryptoquialanine A
Larvae for all experiments were 2nd instars and kept under
controlled temperature conditions (T ¼ 27 ± 1  C), relative hu-
midity (RH ¼ 70 ± 5 %) and photoperiod of 12 h. The bioassay was
performed following the criteria established by Dulmage et al.
(1990), with some modifications. Ten Ae. aegypti larvae were
placed in individual wells of a 6-well cell culture dish containing
10 mL water, 100 mL of liquid feed and the application of extract to
the concentration of 250 mg/mL was added to the larval water; the
negative control receiving only 1 % DMSO (Dimethylsulfoxide) at
the same concentrations, whose mortality did not exceed 10 %.
Survival was monitored for four days checking for the number of
live and dead larvae of each well. Three independent biological
replicates (with three technical replicates each) where performed
for each experiment.
3. Results and discussion
3.1. MSI analysis of green mold on orange fruits surface
MSI has become an important tool for mycologists since it is able
to image thousands of molecules, including metabolites, proteins,
lipids and peptides, providing a visualization of the spatial distri-
bution of secondary metabolites on fungal cultures (Sica et al.,
2014; Buchberger et al., 2018). A few P. digitatum secondary me-
tabolites are described, however, their biological roles in infection
have not yet been fully unveiled (Zhu et al., 2017) as the pathoge-
nicity mechanisms and host specificity (Marcet-Houben et al.,
2012). MSI technique was applied here for the first time to inves-
tigate the metabolites produced by P. digitatum on citrus fruits
surface during infection process and how they are distributed on
the infected fruits, giving initial insights into the role of secondary
metabolites in the infection mechanisms of P. digitatum in the host.
DESI-MSI was applied on oranges infected with P. digitatum at
different stages of the infection process (from 4 to 7 dpi) and an
orange control (Fig. 1). It was possible to detect, in all infection
stages, the tryptoquialanines A and B and their biosynthetic in-
termediates. Fig. 1 shows the MSI signals obtained for ions [MþH]þ
31
at m/z 519.1870, m/z 505.1712, m/z 475.1612, m/z 460.1976 and m/z
459.1664 which correspond respectively to tryptoquialanine A (1)
(C27H26N4O7), tryptoquialanine B (2) (C26H24N4O7), which are
already described to be produced by P. digitatum (Ariza et al., 2002),
tryptoquialanone (4) (C25H22N4O6), 15-dimethyl-2-epi-fumiquina-
zoline A (5) (C25H25N5O4), and deoxytryptoquialanone (6)
(C25H22N4O5) that are intermediates of the tryptoquialanine
biosynthetic pathway. Tryptoquialanines biosynthesis pathway
was well unveiled for Penicillium aethiopicum by Gao et al. (2011)
and intermediates 4, 5 and 6 were also detected for P. digitatum
through MSI analysis, since tryptoquialanine biosynthetic gene
clusters of both fungi show high similarity (Zhu et al., 2017) as show
in Supplementary Fig. S13. All structures (Fig. 2) have been
confirmed through exact masses and typical fragmentation pattern
of tryptoquialanine-like alkaloids. In MS/MS analysis, main typical
fragments were observed at [MþH]þ m/z 156.07, m/z 197.10 and m/z
213.10.
In Fig. 1 it is observed that the alkaloids signals are only present
on the infected surface, indicating that these compounds are pro-
duced during the mycelial growth of P. digitatum. It is also noticed
an accumulation of the metabolites on the fruit surface: the in-
tensity of the signals increases in relation to the time of infection,
especially at 6 dpi, where there is higher concentration level of
these alkaloids in comparison to the previous stages of infection.
Zhu et al. (2017) confirmed through RNA-seq analysis that the
expression of the tryptoquialanine biosynthetic genes occurs since
the first day after inoculation and are in highest level after three
days, indicating a potential biological role of these compounds.
Fig. 1 also shows that the ion m/z 460.1973 relative to the inter-
mediate 5 presents a different spatial distribution compared to the
other analyzed alkaloids being mainly found in the peripheral zone
(white areas). It suggests that 5 is one of the initial intermediates in
the biosynthetic route of tryptoquialanine A (Gao et al., 2011), and
is in higher concentration in the youngest fungal cells.
The detection of the known tryptoquialanines A and B through
MSI shows that this technique can be used for searching new
secondary metabolites. MSI analysis also revealed ions ([MþHþ] m/
z 433.1502 and m/z 503.1920) never before reported from
P. digitatum. These compounds were isolated and investigated by
NMR spectroscopy, since they are not part of the described tryp-
toquialanine biosynthetic pathway.
3.2. Isolation and characterization of secondary metabolites 1, 3
and 7
Secondary metabolites are small organic molecules produced by
various microorganisms and some of them have been shown to
contribute to the pathogenicity of several pathogenic fungi (Scharf
et al., 2014). The discovery of new secondary metabolites produced
by P. digitatum and the understanding of their biological role pro-
vide more information about the infection process and how this
fungus may survive the defense mechanisms of fruit. P. digitatum
extract was obtained from a scaled up experiment and metabolites
of interest were isolated by preparative HPLC. Compounds 3 (m/z
503.1920) and 7 (m/z 433.1503) were detected first by our MSI
technique and were isolated for structural characterization.
Compound 3, previously isolated by Gao et al. (2011), is not part
of the normal biosynthetic pathway of tryptoquialanine A, since it
is a final product formed after the deletion of the TaqE enzyme
responsible for catalyzing a N-hydroxylation and is not yet
described as a natural product. Gao et al. (2011) suggested that the
remaining tailoring steps in the tqa pathway can function in the
absence of N-hydroxylation, leading to 3. So, the detection of 3 in
the wild-type P. digitatum through MSI analysis warranted more
investigation. Compound 3, named as Tryptoquialanine C, was
 32
J.H. Costa et al. / Fungal Biology 123 (2019) 594e600

Fig. 1. (þ) DESI-MSI showing different spacial distributions of molecules over orange peel infected with Penicillium digitatum after 4, 5, 6 and 7 d post inoculation (dpi). (For
interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

isolated from P. digitatum extract and its structure was confirmed
by NMR spectroscopy, confirming that it was a new secondary
metabolite produced by P. digitatum. Supplementary Table S1
shows NMR data obtained for 3. The NMR signals for 3 are
similar to those for 1 (Ariza et al., 2002), with exception of the
presence of an NH signal at 3.05 ppm and the absence of an NOH
signal (dH 7.95) (Supplementary Fig. S12). 1H NMR spectrum
presented three singlets and one doublet of methyl groups, one
methylene group, three methine groups at dH 5.30 (H-2), 5.40 (H-
12) and 6.25 (H-27), and signals corresponding to nine aromatic
protons between dH 7.30 and 8.40. Signals characteristic of an imine
at dc 153.2 (C-26) and of one amide group at dc 161.7 (C-18) sug-
gested that the other aromatic ring was derived from a part bound
to a quinazoline structure. The bonds between the quinazoline

Fig. 2. Structures of 1e6 detected through MSI analysis and of 7e10 identified through GNPS MS/MS database.

J.H. Costa et al. / Fungal Biology 123 (2019) 594e600
structure and the rest of the molecule structure can inferred by
HMBC correlations from dc 153.2 (C-26) to dH 6.25 (H-27) and from
dc 161.7 (C-18) to dH 5.40 (H-12). Long-range HMBC correlations
were also observed from dc 133.9 (C-4) to dH 3.05 (NeH) and from a
lactone carbonyl group at dc 162.5 (C-11) to the singlet at dH 5.30 (H-
2) (Ariza et al., 2002), allowing elucidation of 3.
Through 1H and 13C NMR data (Supplementary Table S2) for 7, it
was possible to conclude that compound 7 is tryptoquivaline L,
alkaloid first isolated from Aspergillus fumigatus and well charac-
terized in the literature (Yamazaki et al., 1979) (Supplementary
Fig. S16 ans S17). Tryptoquivaline L has never have been reported,
until now, as a P. digitatum metabolite in the literature.
Compound 1 was also isolated and its structure was confirmed
through 1H and 13C NMR data (Supplementary Figs. S19 and S20). The
NMR signals obtained for 1 were in agreement with the data already
reported in the literature for this compound (Ariza et al., 2002).
3.3. Evaluation of insecticidal activity of tryptoquialanine A
Zhu et al. (2017) evaluated the involvement of tryptoquialanines
produced P. digitatum in the citrus infection process. The authors
deleted the tqaA gene (Non-ribosomal peptide synthetase)
responsible for tryptoquialanine A production. Phenotype assays
suggested no significant variation between the non-TQA mutant
and the wild-type strain in fungal growth, concluding that tryp-
toquialanine A was not involved in the pathogenicity of P. digitatum
and does not act as a virulence factor, implying a different biological
role for the indole alkaloids (Zhu et al., 2017).
Based on structure similarity, tryptoquialanine A (1) is consid-
ered a tremorgenic mycotoxin (Ariza et al., 2002) acting on the
central nervous system of vertebrate animals, causing symptoms
such as decreased activity and immobility (Gao et al., 2011). Since 1
is not related to the P. digitatum pathogenicity, Zhu et al. (2017)
suggests that it may have an insecticidal biological function
within its micro ecosystem.
Compound 1 was isolated and applied in insecticidal assays to
evaluate the hypothesis suggested by Zhu et al. (2017). In 24 h after
the exposure to 1, the mortality rate of Ae. Aegypti larvae was 37 %.
The larvae mortality gradually increased during the days and
reached 81 % on the fourth day (Fig. 3). This result indicates a very
pronounced insecticidal activity of 1 and suggests that 1 may be
important as a biocontrol against insects during orange decay.
3.4. Comparing alkaloids in different citrus: molecular network
Although DESI-MSI is a great tool to discover new secondary
metabolites involved in biological interactions, the technique is
Fig. 3. Mean and standard deviation of the mortality rate of Aedes aegypti larvae for
96 h. The differences between Tryptoquialanine A treated samples and PBS control
samples were evaluated by unpaired t test (**, p < 0.05, ***p < 0.05, ****p < 0.0001),
which had significant effects on mortality.
33
limited to the analysis of the fruits surface, so that internal me-
tabolites remain undetected (Sica et al., 2014). Other complemen-
tary approaches are needed to further characterize the metabolic
profile during infection. Therefore, molecular networking was used
here, as a complement to DESI-IMS, to annotate secondary me-
tabolites produced during P. digitatum's infection.
Another alternative to identify unknown compounds, involves
metabolite fragmentation patterns acquired through tandem MS
analysis, that can be matched to those presented in databases such
as GNPS, PubChem and others useful databases for dereplication
(Covington et al., 2017; Smith et al., 2005). Nowadays, the largest
natural product public database with MS/MS spectra is GNPS:
Global Natural Products Social Molecular Networking, containing
more than 140.000 natural products (Covington et al., 2017;
Gaude ^ncio and Pereira, 2015).
In this paper, we further investigated the metabolic profile of
P. digitatum during infection in different citrus hosts. As mentioned
before, this fungus is considered a major pathogen of citrus fruit
(Ghooshkhaneh et al., 2018). The metabolic profiles of P. digitatum
in distinct citrus fruits (sicilian lemon, tangerine and orange), or-
ange juice and PDA were analyzed in the GNPS database through
tandem MS (MS/MS) aimed at finding new metabolites.
After generating the molecular networks, the node connectivity
was visualized (Supplementary Fig. S22). Each node represents one
MS/MS spectrum and is labeled with the parent (precursor) mass.
Since fragmentation spectra generally reflect the chemical struc-
tures of the fragmented ions, it becomes possible to represent the
produced metabolites in clusters of similar structures through the
GNPS platform (Nguyen et al., 2013) (as observed in Fig. 4). For
P. digitatum extracts the network contained 31 different clusters.
The visualization of cluster B showed that the ions [MþH]þ m/z
519.1871 and 505.1712, clustered together in the molecular network
(Fig. 4), are tryptoquialanine A 1 and tryptoquialanine B 2 respec-
tively, major secondary metabolites for P. digitatum (Ariza et al.,
2002), and observed in all growth media. Interestingly, 1 and 2
were observed in vivo in all citrus fruits, suggesting an important
biological role, which was confirmed here through the insecticidal
biossays.
The tryptoquialanine intermediates detected by MSI analysis
were also observed in P. digitatum extracts. 5 (m/z 460.1976) is one
of the precursors of tryptoquialanine pathway and was observed in
orange and PDA media. Similarly, 6 (m/z 459.1664) and 3 (m/z
503.1920) were also observed, being the last one found in all
growth media and 6 was only produced in sicilian lemon and PDA
media. Finally, 4 (m/z 475.1612), another intermediate involved in
the tryptoquialanine biosynthesis, was only detected in PDA under
the cultivation conditions tested.
In addition, cluster A presented some well-known compounds
in the GNPS database bolded by red (Fig. 4). These compounds were
identified as tryptoquivaline L (7), tryptoquivaline Q (8), fumiqui-
nazoline A (9) and fumiquinazoline C (10), with exact mass of
[MþH]þ m/z 433.1503, [M-H2OþH]þ m/z 417.1555 [M-H2OþH]þ m/z
428.1717 and [MþH]þ m/z 444.1665, respectively. All structures are
represented in Fig. 2 and are secondary metabolites for the first
time described in P. digitatum further confirming molecular
networking as a great tool for natural products discovery.
Tryptoquivaline L (7) was detected in PDA media by GNPS in
agreement with the MSI analyses. Both techniques were able to
identify 7 as a secondary metabolite produced by P. digitatum, and
these results were confirmed by the isolation and structure eluci-
dation of 7. Tryptoquivaline L has been previously reported for the
fungus A. fumigatus (Yamazaki et al., 1978, 1979), and Neosartorya
siamenisis KUFC 6349 (Buttachon et al., 2012). This compound was
also isolated from marine sponge associated fungus Neosartorya
paulistensis KUFC 7897 (Gomes et al., 2014) and the marine-derived

J.H. Costa et al. / Fungal Biology 123 (2019) 594e600
Fig. 4. Zoom in of MS/MS network analysis (Supplementary Fig. S22) of PDA, Orange, Sicilian Lemon, Orange juice and Tangerine extracts from P. digitatum. Colors indicated in the
key correspond to the different growth media source. Nodes bolded by black lines represents the compounds observed in the MSI analysis and nodes bolded by red represents GNPS
hits. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
fungus Neosartorya laciniosa KUFC 7896 (Eamvijarn et al., 2013), as
well as Neosartorya takakii KUFC 7898 (Zin et al., 2015). Trypto-
quivaline Q (8) has been previously isolated from a marine-derived
fungus Neosartorya sp. HN-M-3 (Sun et al., 2012). Regarding the
fumiquinazolines A (9) and C (10), Numata et al., (1992) isolated
both molecules from the mycelium of A. fumigatus. These com-
pounds exhibited moderate cytotoxicity against the cultured P-388
lymphocytic leukemia cells (Numata et al., 1992). Fumiquinazolines
and tryptoquialanines present similar enzymes in their biosyn-
thetic pathway and also have Fumiquinazoline F as an intermediate
in common (Gao et al., 2011; Ames et al., 2011). Sequence analysis
of TqaA enzyme from P. aethiopicum tryptoquialanine pathway
revealed high homology with Af12080 enzyme from A. fumigatus
fumiquinazoline pathway. TqAa enzyme, in the tryptoquialanine
pathway, catalyzes the biosynthesis of the Fumiquinazoline F, in-
termediate that can be used in both pathways (Gao et al., 2011).
Therefore, is not surprising to detect fumiquinazolines in
P. digitatum's extract when we analyze both biosynthetic pathways.
Until now, 8, 9 and 10 have never been reported as P. digitatum
secondary metabolites.
GNPS molecular networking allowed consistent identification of
metabolites present not only on the surface of the fruit but also in
the different P. digitatum extracts. The detection of ions through
MSI and GNPS analyses confirm the production of known and new
indole alkaloids in the mycelium of P. digitatum.
4. Conclusions
Despite great economic interest related to P. digitatum, studies
concerning this pathogen-host system have mainly focused on
treatments against infection symptoms. The specificity concerning
the host as well as the biological roles of secondary metabolites
involved in the pathogen-host interaction remain largely unknown.
In this context, MSI has emerged as a powerful tool to gain more
insights concerning the natural products produced in P. digitatum
infection that may have an important biological role. The MSI
analysis enabled monitoring and understanding the indole alkaloid
production by P. digitatum during infection process in citrus.
Complementarily to MSI, GNPS molecular networking allowed the
34
study of the metabolic profile produced by P. digitatum in different
citrus hosts, confirming the production of known and new alkaloids
in the mycelium of P. digitatum. Therefore, MSI and Molecular
Networking taken together, provide a solid insight into different
nature interactions and pathogen infection process. Through these
complementary approaches it was possible to provide the first in-
sights about the indole alkaloids involved in the P. digitatum-citrus
interaction and initial thoughts of the possible biological role
associated to them. In addition, the knowledge concerning the
potential biological role of tryptoquialanines can serve as a target to
the development of new biological insecticides to be potentially
used in agriculture. In conclusion, the secondary metabolism of
P. digitatum and the roles of natural products in the infection are
still an open field for new research opportunities.
Acknowledgements
This study was financed in part by the Coordenaça ~o de Aper-
feiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance
Code 001, Fundaç~ao de Amparo a Pesquisa no Estado de Sa ~o Paulo
[grant number FAPESP 2017/24462-4 and 2018/03670-0] and Pro -
Reitoria de Pesquisa Unicamp (PRP/FAEPEX). MCFP is recipient of a
PhD Fellowship from CAPES (grant number 8887.137194/2017-00).
JASN is recipient of a FAPESP Young Investigator Award (grant
number 2013/11343-6).
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.funbio.2019.03.002.
References
Ames, B.D., Haynes, S.W., Gao, X., Evans, B.S., Kelleher, N.L., Tang, Y., Walsh, C.T.,
2011. Complexity generation in fungal peptidyl alkaloid biosynthesis: oxidation
of fumiquinazoline a to the heptacyclic hemiaminal fumiquinazoline C by the
flavoenzyme Af12070 from Aspergillus fumigatus. Biochemistry 50 (4), 8756-
8787-69. http://doi.org/10.1021/bi201302w.
Angolini, C.F.F., Vendramini, P.H., Araújo, F.D.S., Araújo, W.L., Augusti, R.,
Eberlin, M.N., De Oliveira, L.G., 2015. Direct protocol for ambient mass

J.H. Costa et al. / Fungal Biology 123 (2019) 594e600
spectrometry imaging on agar culture. Anal. Chem. 87, 6925e6930. https://doi.
org/10.1021/acs.analchem.5b01538.
Ariza, M.R., Larsen, T.O., Petersen, B.O., Duus, J.Ø., Barrero, A.F., 2002. Penicillium
digitatum metabolites on synthetic media and citrus fruits. J. Agric. Food Chem.
50, 6361e6365. https://doi.org/10.1021/jf020398d.
Barmore, C.R., Brown, G.E., 1981. Polygalacturonase from citrus fruit infected with
Penicillium italicum. Phytopathology 71, 328e331.
Buchberger, A.R., DeLaney, K., Johnson, K., Li, L., 2018. Mass spectrometry imaging: a
review of emerging advancements and future insights. Anal. Chem. 90,
240e265. http://doi.org/10.1021/acs.analchem.7b04733.
Buttachon, S., Chandrapatya, A., Manoch, L., Silva, A., Gales, L., Bruye re, C., Kijjoa, A.,
2012. Sartorymensin, a new indole alkaloid, and new analogues of tryptoqui-
valine and fiscalins produced by Neosartorya siamensis (KUFC 6349). Tetrahe-
dron 68, 3253e3262.
Covington, B.C., McLean, J.A., Bachmann, B.O., 2017. Comparative mass
spectrometry-based metabolomics strategies for the investigation of microbial
secondary metabolites. Nat. Prod. Rep. 34, 6e24. https://doi.org/10.1039/
C6NP00048G.
Dulmage, H.T., Yousten, A.A., Singer, S., Lacey, L.A., 1990. Guidelines for Production
of Bacillus Thuringiensis H-14 and Bacillus Sphaericus. UNDP/World Bank/WHO,
Steering Committee to Biological Control of Vetores, Geneva, p. 59.
Eamvijarn, A., Gomes, N.M., Dethoup, T., Buaruang, J., Manoch, L., Silva, A., Pedro, M.,
Marini, I., Roussis, V., Kijjoa, A., 2013. Bioactive meroditerpenes and indole al-
kaloids from the soil fungus Neosartorya fischeri (KUFC 6344), and the marine-
derived fungi Neosartorya laciniosa (KUFC 7896) and Neosartorya tsunodae
(KUFC 9213). Tetrahedron 69, 8583e8591.
Gao, X., Chooi, Y.H., Ames, B.D., Wang, P., Walsh, C.T., Tang, Y., 2011. Fungal indole
alkaloid biosynthesis: genetic and biochemical investigation of the trypto-
quialanine pathway in Penicillium aethiopicum. J. Am. Chem. Soc. 133 (8),
2729e2741. https://doi.org/10.1021/ja1101085.
Gaude ^ncio, S.P., Pereira, F., 2015. Dereplication: racing to speed up the natural
products discovery process. Nat. Prod. Rep. 32, 779e810.
Gomes, N.M., Bessa, L.J., Buttachon, S., Costa, P.M., Buaruang, J., Dethoup, T., Silva, A.,
Kijjoa, A., 2014. Antibacterial and antibiofilm activities of tryptoquivalines and
meroditerpenes isolated from the marine-derived fungi Neosartorya paulis-
tensis, N. laciniosa, N. tsunodae, and the soil fungi N. fischeri and N. siamensis.
Mar. Drugs 12, 822e839.
Ghooshkhaneh, N.G., Golzarian, M.R., Mamarabadi, M., 2018. Detection and classifi-
cation of citrus green mold caused by Penicillium digitatum using multispectral
imaging. J. Sci. Food Agric. 98, 3542e3550. https://doi.org/10.1002/jsfa.8865.
Lei, Z., Huhman, D., Sumner, L.W., 2011. Mass spectrometry strategies in metab-
olomics. J. Biol. Chem. jbc-R111 https://doi.org/10.1074/jbc.R111.238691.
Lopez-Pe rez, M., Ballester, A., Gonz alez-Candelas, L., 2015. Identification and func-
tional analysis of Penicillium digitatum genes putatively involved in virulence
towards citrus fruit. Mol. Plant Pathol. 16, 262e275. https://doi.org/10.1111/
mpp.12179.
Macarisin, D., Cohen, L., Eick, A., Rafael, G., Belausov, E., Wisniewski, M., Droby, S.,
2007. Penicillium digitatum suppresses production of hydrogen peroxide in
host tissue during infection of citrus fruit. Phytopathology 97, 1491e1500.
https://doi.org/10.1094/PHYTO-97-11-1491.
Marcet-Houben, M., Ballester, A.R., La Fuente, B., Harries, E., Marcos, J.F., Gonza lez-
Candelas, L., Gabaldo  n, T., 2012. Genome sequence of the necrotrophic fungus
35
Penicillium digitatum, the main postharvest pathogen of citrus. BMC Genomics
13, 646. https://doi.org/10.1186/1471-2164-13-646.
Nguyen, D.D., Wu, C.H., Moree, W.J., Lamsa, A., Medema, M.H., Zhao, X.,
Gavilan, R.G., Aparicio, M., Atencio, L., Jackson, C., Ballesteros, J., Sanchez, J.,
Watrous, J.D., Phelan, V.V., van de Wien, C., Kersten, R.D., Mehnaz, S., De Mot, R.,
Shank, E.A., Charusanti, P., Nagarajan, H., Duggan, B.M., Moore, B.S., Bandeira, N.,
Palsson, B.O., Pogliano, K., Gutie rrez, M., Dorrestein, P.C., 2013. MS/MS
networking guided analysis of molecule and gene clusters families. Proc. Natl.
Acad. Sci. U. S. A. 110, E2611eE2620. https://doi.org/10.1073/pnas.1303471110.
Numata, A., Takahashi, C., Matsushita, T., Miyamoto, M., Kawai, K., Usami, Y.,
Matsumura, E., Inoue, M., Ohishi, H., Shingu, T., 1992. Fumiquinazolines, novel
metabolites of a fungus isolates from a saltfish. Tetrahedron Lett. 33 (12),
1621e1624. https://doi.org/10.1016/S0040-4039(00)91690-3.
Panebianco, S., Vitale, A., Platania, C., Restuccia, C., Polizzi, G., Cirvilleri, G., 2014.
Postharvest efficacy of resistance inducers for the control of green mold on
important Sicilian citrus varieties. J. Plant Dis. Prot. 121, 177e183. https://doi.
org/10.1007/BF03356507.
Perez, M.F., Ibarreche, J.P., Isas, A.S., Sepulveda, M., Ramallo, J., Dib, J.R., 2017.
Antagonistic yeasts for the biological control of Penicillium digitatum on
lemons stored under export conditions. Biol. Control 115, 135e140. https://doi.
org/10.1016/j.biocontrol.2017.10.006.
Prusky, D., McEvoy, J.L., Saftner, R., Conway, W.S., Jones, R., 2004. Relationship be-
tween host acidification and virulence of Penicillium spp. on apple and citrus
fruit. Phytopathology 94, 44e51. http://doi.org/10.1094/PHYTO.2004.94.1.44.
Scharf, D.H., Heinekamp, T., Brakhage, A.A., 2014. Human and plant fungal patho-
gens: the role of secondary metabolites. PLoS Pathog. 10 (1), e1003859. https://
doi.org/10.1371/journal.ppat.1003859.
Sica, V.P., Raja, H.A., El-Elimat, T., Oberlies, N.H., 2014. Mass spectrometry imaging of
secondary metabolites directly on fungal cultures. RSC Adv. 4, 63221e63227.
https://doi.org/10.10139/c4ra11564c.
Smith, C.A., O'Maille, G., Want, E.J., Qin, C., Trauger, S.A., Brandon, T.R., Siuzdak, G.,
2005. METLIN: a metabolite mass spectral database. Ther. Drug Monit. 27,
747e751.
Sun, F., Chen, G., Bai, J., Pei, Y., 2012. Two new alkaloids from a marine-derived
fungus Neosartorya sp. HN-M-3. J. Asian Nat. Prod. Res. 14 (12), 1109e1115.
Yamazaki, M., Fujimoto, H., Okuyama, E., 1978. Structure determination of six fungal
metabolites, tryptoquivaline E, F, G, H, I and J from Aspergillus fumigatus. Chem.
Pharm. Bull. 26, 111e117.
Yamazaki, M., Okuyama, E., Maebayashi, Y., 1979. Isolation of some new
tryptoquivaline-related metabolites from Aspergillus fumigatus. Chem. Pharm.
Bull. 27, 1611e1617.
Zhang, T., Sun, X., Xu, Q., Candelas, L.G., Li, H., 2013. The pH signaling transcription
factor PacC is required for full virulence in Penicillium digitatum. Appl. Micro-
biol. Biotechnol. 97, 9087e9098. https://doi.org./10.1007/s00253-013-5129-x.
Zhu, C., Sheng, D., Wu, X., Wang, M., Hu, X., Li, H., Yu, D., 2017. Identification of
secondary metabolite biosynthetic gene clusters associated with the infection
of citrus fruit by Penicillium digitatum. Postharvest Biol. Technol. 134, 17e21.
https://doi.org/10.1016/j.postharvbio.2017.07.011.
Zin, W.W.M., Buttachon, S., Buaruang, J., Gales, L., Pereira, J.A., Pinto, M.M.M.,
Silva, A.M.S., Kijjoa, A., 2015. A new meroditerpene and a new tryptoquivaline
analog from the algicolous fungus Neosartorya takakii KUFC 7898. Mar. Drugs
13 (6), 3776e3790. https://doi.org/10.3390/md13063776.
 36

3.3 Informações suplementares do capítulo I

Monitoring indole alkaloid production by Penicillium digitatum during infection

process in citrus by Mass Spectrometry Imaging and Molecular Networking

Jonas Henrique Costaa, Jaqueline Moraes Baziolia,b, Eder de Vilhena Araújoa, Pedro
Henrique Vendraminia, Mariana Cristina de Freitas Portoc, Jayme A. Souza-Netoc,
Taícia Pacheco Filla*

a Institute of Chemistry, Universidade Estadual de Campinas, CP 6154, 13083-970,

Campinas – SP, Brazil

b Faculty of Pharmaceutical Sciences, Universidade Estadual de Campinas, 13083-

859 Campinas – SP, Brazil

C São Paulo State University (UNESP), School of Agricultural Sciences, Department of

Bioprocesses and Biotechnology, Central Multiuser Laboratory, Botucatu, Brazil

*Corresponding author: phone +55-19-35213092, e-mail taicia@unicamp.br

S1. MSI analysis of green mold on orange fruits

Fig. S1. Oranges infected with P. digitatum used for MSI analysis.

Fig. S2. Samples of infected orange peels analyzed by MSI technique.

 37

Fig. S3. Photograph of DESI source on sample of orange peel infected with P. digitatum.

Fig. S4. DESI-MS (+) of compound 1.

Fig. S5. DESI-MS (+) of compound 2.

 38

Fig. S6. DESI-MS (+) of compound 3.

Fig. S7. DESI-MS (+) of compound4.

Fig. S8. DESI-MS (+) of compound 5.

 39

Fig. S9. DESI-MS (+) of compound 6.

Fig. S10. DESI-MS (+) of compound 7.

Fig. S11. Comparison of tryptoquialanine biosynthetic gene clusters between P. digitatum and P.
aethiopicum showing high similarity. Graphic adapted from antiSMASH.
(https://fungismash.secondarymetabolites.org/#!/start)
 40

S2. Isolation and characterization of secondary metabolites

Table S1. 1H, HSQC 13C and HMBC NMR data for 3 (500 MHz, CDCl3) (δ in ppm, J in Hz).
3

Position 1
H δ (m, J) HSQC 13C δ HMBC
2 5.30 (s) 81.3 -
3 - 78.4 -
4 - 133.9 H-6, H-8, NH-16
5 7.79 (dd, 7.9, 1.2) 127.8 H-7
6 7.52 (td, 7.9,1.2) 123.8 -
7 7.59 (m) 128.1 H-5, H-8
8 7.67 (d, 8.1) 116.3 H-6, H-7
9 - 138.7 H-7
11 - 162.5 H-2
12 5.40 t (9.9) 55.1 H-18
13 (a) 3.20 (dd, 13.3, 9.7) 33.5 H-12
13 (b) 2.97 (dd, 13.3, 10) -
14 - 175.9 H-29, H-30, H-2
15 - 66.1 H-29, H-30
18 - 161.7 H-12
19 - 120.8 H-21, H-22
20 8.35 (dd, 8.5, 1.1) 127.3 H-22
21 7.56 (td, 7.7, 1.2) 132.3 H-23
22 7.84 (td, 7.9, 1.2) 135.3 H-20
23 7.39 (dd, 7.2, 1.1) 125.8 H-21
24 - 149.2 -
26 - 153.2 H-28
27 6.25 (q, 6.3) 69.1 H-28
28 1.83 (d, 6.4) 18.6 H-27
29 1.57 (s) 26.6 H-30
30 1.56 (s) 25.4 H-29
31 - 171.0 H-32
32 2.21(s) 21.0 H-31
NH 3.05 (d, 5.2) - -
 42

Fig. S14. 1H-13C HMBC NMR spectrum of 3 (500.13 MHz, CDCl3).

Fig. S15. HPLC-MS/MS analysis of compound 3. A) total ion chromatogram, B) mass spectrum of ion
[M+H]+ m/z 503.1920 and C) MS/MS spectrum of ion [M+H]+ m/z 503.1920.
 43

Table S2. 1H and 13C NMR data for 7 (500 MHz, DMSO-d6) (δ in ppm, J in Hz).

Position 1
H δ (m, J) Cδ
13

2 5.40 (s) 84.9

3 - 80.9
4 - 134.9
5 7.95 (d,7.8) 126.0
6 7.29 (dd,7.8, 7.6) 125.8
7 7.45 (dd,7.8, 7.6) 131.9
8 7.78 (d, 7.6) 113.9
9 - 138.2
11 - 171.8
12 6.10 t (10.2) 56.4
13 (a) 2.85 (m) 33.1
13 (b) 3.02 (m)
14 - 172.2
15 - 65.8
18 - 158.9
19 - 123.8
20 8.01 (d, 8.0) 126.3
21 7.49 (dd,8.0, 7.8) 128.3
22 7.72(dd,8.0, 7.6) 136.4
23 7.23 (d, 8.0) 127.9
24 - 150.4
26 8.66 (s) 149.5
27 1.38 (s, 6.3) 14.9
28 1.29 (s, 6.4) 25.3
OH 8.73 (s) -
 44

Fig. S16. 1H NMR spectrum of 7 (500.13 MHz, DMSO-d6)

Fig. S17. 13C NMR spectrum of 7 (125.7 MHz, DMSO-d6)

Fig. S18. HPLC-MS/MS analysis of compound 7. A) total ion chromatogram, B) mass spectrum of ion
[M+H]+ m/z 433.1503 and C) MS/MS spectrum of ion [M+H]+ m/z 433.1503
 46

Fig. S21. HPLC-MS/MS analysis of compound 1 A) total ion chromatogram, B) mass spectrum of ion
[M+H]+ m/z 519.1871 and C) MS/MS spectrum of ion [M+H]+ m/z 519.1871.

S4. Comparing alkaloids in different growing media: Molecular Network

Fig. S22. Complete MS/MS network analysis of P. digitatum extracts. Compounds with similar
fragmentation patterns form clusters in the network. The tryptoquialanines and intermediates are
grouped in clusters A and B.
 47

Fig. S23. MS/MS spectrum for compound 2.

Fig. S24. MS/MS spectrum for compound 4.

Fig. S25. MS/MS spectrum for compound 5.

Fig. S26. MS/MS spectrum for compound 6.

 48

Fig. S27. MS/MS match between GNPS database (green) and P. digitatum extract (black) for
compound 7.

Fig. S28. Mass spectrum of ions [M-H2O+H]+ and [M+H]+ for compound 8.

Fig. S29. MS/MS match between GNPS database (green) and P. digitatum extract (black) for
compound 8.
 49

Fig. S30. Mass spectrum of ions [M-H2O+H]+ and [M+H]+ for compound 9.

Fig. S31. MS/MS match between GNPS database (green) and P. digitatum extract (black) for
compound 9.

Fig. S32. Mass spectrum of ions [M-H2O+H]+ and [M+H]+ for compound 10.
 50

Fig. S33. MS/MS match between GNPS database (green) and P. digitatum extract (black) for
compound 10.
 51

4 CAPÍTULO II
Triptoquialaninas fitotóxicas produzidas in vivo por Penicillium digitatum são
exportadas em vesiculas extracelulares

O conteúdo deste capítulo é composto pelo artigo intitulado “Phytotoxic

tryptoquialanines produced in vivo by Penicillium digitatum are exported in
extracellular vesicles”, publicado no periódico mBio. A reprodução pelos autores
é autorizada sob os termos da Creative Commons Attribution 4.0 license (anexo
10.2).

Referência: Costa, J.H., Bazioli, J.M., Barbosa, L.D., et al. 2021. Phytotoxic
Tryptoquialanines Produced In Vivo by Penicillium digitatum Are Exported in
Extracellular Vesicles. mBio, v. 12(1), e03393-20. doi: 10.1128/mbio.03393-20

Versão final publicada disponível em:

https://mbio.asm.org/content/12/1/e03393-20

4.1 Resumo
Neste trabalho investigou-se o envolvimento das triptoquialaninas nos
danos causados por P. digitatum às frutas cítricas. Ensaios de germinação utilizando
triptoquialanina A e sementes de Citrus sinensis foram feitos. As sementes expostas
à uma concentração de 3.000 ppm do alcaloide tiveram inibição completa de
germinação e alteração metabólica. Considerando o efeito fitotóxico das
triptoquialaninas, o mecanismo de transporte no qual esses alcaloides são exportados
a partir das células de P. digitatum foi investigado. Pela primeira vez foram isoladas e
caracterizadas vesículas extracelulares (VEs) secretadas por P. digitatum in vivo e in
vitro. A análise do conteúdo das VEs confirmou a presença das triptoquialaninas e de
micotoxinas do tipo fungisporin. Novos ensaios fitotóxicos utilizando as VEs isoladas
de P. digitatum também mostraram uma ação fitotóxica das VEs através de danos nos
tecidos das sementes. Ainda, através das redes moleculares foi possível observar a
presença dos metabólitos de P. digitatum também em novos hospedeiros reportados
na literatura (ameixa). Os resultados apresentados revelam um papel fitopatogênico
das triptoquialaninas e VEs produzidas por P. digitatum.
 60
52
4.2
4.2 Artigo:
Artigo: “Phytotoxic
"Phytotoxic tryptoquialanines
tryptoquialanines produced
produced in
in vivo
vivo by
by Penicillium
Penicillium digitatum
digitatum are
are exported
RESEARCHin
exported in
ARTICLE
extracellular vesicles”
extracellular vesicles" Host-Microbe Biology

Phytotoxic Tryptoquialanines Produced In Vivo by Penicillium

digitatum Are Exported in Extracellular Vesicles
Jonas Henrique Costa,a Jaqueline Moraes Bazioli,a,b Luidy Darllan Barbosa,a Pedro Luis Theodoro dos Santos Júnior,a
Flavia C. G. Reis,c,d Tabata Klimeck,c Camila Manoel Crnkovic,e Roberto G. S. Berlinck,f Alessandra Sussulini,a
Marcio L. Rodrigues,c,g Taícia Pacheco Filla

Downloaded from http://mbio.asm.org/ on February 9, 2021 by guest

a Institute of Chemistry, University of Campinas, CP 6154, Campinas, São Paulo, Brazil
b Faculty of Pharmaceutical Sciences, University of Campinas, Campinas, São Paulo, Brazil
c Instituto Carlos Chagas, Fundação Oswaldo Cruz (Fiocruz), Curitiba, Paraná, Brazil
d Centro de Desenvolvimento Tecnológico em Saúde (CDTS), Fiocruz, Rio de Janeiro, Brazil
e Department of Biochemical and Pharmaceutical Technology, School of Pharmaceutical Sciences, University of São Paulo, São Paulo, São Paulo, Brazil
Instituto de Química de São Carlos, Universidade de São Paulo, CP 780, São Carlos, São Paulo, Brazil
f

g Instituto de Microbiologia Paulo de Góes (IMPG), Universidade Federal do Rio de Janeiro, Rio de Janeiro, Rio de Janeiro, Brazil

ABSTRACT Penicillium digitatum is the most aggressive pathogen of citrus fruits.

Tryptoquialanines are major indole alkaloids produced by P. digitatum. It is
unknown if tryptoquialanines are involved in the damage of citrus fruits caused
by P. digitatum. To investigate the pathogenic roles of tryptoquialanines, we ini-
tially asked if tryptoquialanines could affect the germination of Citrus sinensis
seeds. Exposure of the citrus seeds to tryptoquialanine A resulted in a complete
inhibition of germination and an altered metabolic response. Since this phyto-
toxic effect requires the extracellular export of tryptoquialanine A, we investi-
gated the mechanisms of extracellular delivery of this alkaloid in P. digitatum. We
detected extracellular vesicles (EVs) released by P. digitatum both in culture and
during infection of citrus fruits. Compositional analysis of EVs produced during
infection revealed the presence of a complex cargo, which included tryptoquiala-
nines and the mycotoxin fungisporin. The EVs also presented phytotoxicity activity
Citation Costa JH, Bazioli JM, Barbosa LD, dos
in vitro and caused damage to the tissues of citrus seeds. Through molecular network- Santos Júnior PLT, Reis FCG, Klimeck T, Crnkovic
ing, it was observed that the metabolites present in the P. digitatum EVs are produced CM, Berlinck RGS, Sussulini A, Rodrigues ML, Fill
in all of its possible hosts. Our results reveal a novel phytopathogenic role of P. digita- TP. 2021. Phytotoxic tryptoquialanines
produced in vivo by Penicillium digitatum are
tum EVs and tryptoquialanine A, implying that this alkaloid is exported in EVs during exported in extracellular vesicles. mBio 12:
plant infection. e03393-20. https://doi.org/10.1128/mBio
.03393-20.
IMPORTANCE During the postharvest period, citrus fruits can be affected by phyto-
Editor James W. Kronstad, University of British
pathogens such as Penicillium digitatum, which causes green mold disease and is re- Columbia
sponsible for up to 90% of total citrus losses. Chemical fungicides are widely used to Copyright © 2021 Costa et al. This is an open-
prevent green mold disease, leading to concerns about environmental and health access article distributed under the terms of
the Creative Commons Attribution 4.0
risks. To develop safer alternatives to control phytopathogens, it is necessary to under- International license.
stand the molecular basis of infection during the host-pathogen interaction. In the P. Address correspondence to Marcio L.
digitatum model, the virulence strategies are poorly known. Here, we describe the pro- Rodrigues, marcio.rodrigues@fiocruz.br, or
Taícia Pacheco Fill, taicia@unicamp.br.
duction of phytotoxic extracellular vesicles (EVs) by P. digitatum during the infection
This article is a direct contribution from Marcio
of citrus fruits. We also characterized the secondary metabolites in the cargo of EVs L. Rodrigues, a Fellow of the American
and found in this set of molecules an inhibitor of seed germination. Since EVs and sec- Academy of Microbiology, who arranged for
and secured reviews by Aaron Mitchell,
ondary metabolites have been related to virulence mechanisms in other host-patho-
University of Georgia, and Maysa Furlan,
gen interactions, our data are important for the comprehension of how P. digitatum Universidade Estadual Paulista.
causes damage to its primary hosts. Received 2 December 2020
Accepted 16 December 2020
KEYWORDS fungi, extracellular vesicles, herbicidal activity, host-pathogen interaction, Published 9 February 2021
P. digitatum, tryptoquialanines, Penicillium digitatum
®
January/February 2021 Volume 12 Issue 1 e03393-20 mbio.asm.org 1
 53
®
Costa et al.

C itriculture is a worldwide multi-billion-dollar activity (1). Brazil, China, and the

United States are the major producers of citrus (2). In Brazil, 230,000 direct and
indirect jobs are related to citriculture (3). The citrus industry in Brazil corresponded to
US$6.5 billion in revenues in 2019 (3). Citrus fruits can be affected by different diseases,
leading to up to 50% of fruit losses and causing a negative economic impact (4–6).
Diseases caused by fungal pathogens are the most adverse factors causing fresh
fruit and vegetable losses during the postharvest period (7, 8). Postharvest losses due
to fungal diseases can reach 30 to 55% of production (8–10). The most damaging post-
harvest disease in citrus is the green mold caused by Penicillium digitatum, which
accounts for up to 90% of citrus losses (4–6).
Demethylation inhibitors (DMI), including prochloraz and imazalil, are fungicides
used to combat P. digitatum (4). However, this practice has raised concerns about its

Downloaded from http://mbio.asm.org/ on February 9, 2021 by guest

effects on human health and the development of antifungal resistance (10, 11).
Therefore, developing safer approaches to control postharvest diseases has become a
global trend (7, 8, 10, 11). The development of alternative antifungal tools demands an
improved knowledge of how P. digitatum causes damage to citrus fruits. Most efforts
in this direction have focused on the use of biocontrol agents, antagonist microorgan-
isms, and natural products (8, 11, 12) to neutralize virulence factors (4, 13). Nevertheless,
the molecular mechanisms underlying the induction of P. digitatum-mediated damage
in host cells remain poorly known (4, 5, 14).
Secondary metabolites were reported as essential for fungal pathogenicity and a
consequent attenuation of the plant defense responses (4, 5, 14). Siderophores, for
instance, have been associated with fungal virulence by iron sequestration (4, 15).
Secondary metabolites have not yet been associated with the pathogenic process pro-
moted by P. digitatum (4). Tryptoquialanines are major metabolites produced by P. dig-
itatum (16). However, their role in P. digitatum phytopathogenicity is unclear, even
though these compounds displayed insecticidal (17) and antifungal (18) activities.
In addition to secondary metabolites, extracellular vesicles (EVs) have been associ-
ated with the pathogenesis of several infectious diseases (19, 20). EVs are spherical
structures that are released by bacteria, fungi, and plant cells (19–21). EVs are delimited
by a lipid bilayer membrane in association with proteins, lipids, enzymes, pigments,
polysaccharides, and RNAs (19–21). In fungi, EVs were first reported in the human-path-
ogenic yeast Cryptococcus neoformans and subsequently characterized in several fun-
gal pathogens (20, 21). In host-microbe interactions, EVs are key players determining
the pathogenic outcome (22, 23) and mediating the transfer of virulence traits (24). In
plant infections, the roles of EVs have been superficially explored, and the knowledge
of how EVs impact the plant physiology is limited (22, 23). It is known that under stress
conditions, release of EVs by plants is increased in response to infection (22, 23). EVs
are involved in the defense of plants against pathogens, forming physical barriers or
delivering molecules that are toxic to invading microbes (22, 23). On the other hand,
pathogen EVs can inhibit plant immune responses through the export of virulence fac-
tors (22). So far, no information on the production of EVs by phytopathogens and their
association with secondary metabolite cargo is available.
Here, we report the phytotoxic activity of indole alkaloids and EVs produced by P.
digitatum in germination assays of Citrus sinensis seeds. The export of tryptoquialanines
in P. digitatum involved EVs. Untargeted metabolomics was applied to confirm the EV-
mediated metabolite export in the possible hosts (citrus and stone fruits) affected by P.
digitatum.

RESULTS
Phytotoxicity activity of tryptoquialanine. To investigate the pathogenic poten-
tial of tryptoquialanine A (TA), we first purified this alkaloid from P. digitatum’s crude
extracts (Fig. S1). We then evaluated its phytotoxicity in germination assays of Citrus
sinensis seeds. TA significantly inhibited seed germination at all tested concentrations.
Seeds exposed to concentrations of TA lower than 1,000 ppm exhibited a delay in their

January/February 2021 Volume 12 Issue 1 e03393-20 mbio.asm.org 2

 54
®
Extracellular Vesicles produced by P. digitatum

Downloaded from http://mbio.asm.org/ on February 9, 2021 by guest

FIG 1 Tryptoquialanine A (TA) is an efficient inhibitor of germination in C. sinensis seeds. (A) Untreated
seeds (negative control) showed a regular pattern of germination. (B to F) Seeds exposed to TA (500,
1,000, or 3,000 ppm; panels B, C, and D, respectively) or to the commercial herbicide Roundup (3,000 or
10,000 ppm; panels E and F, respectively) manifested defective germination. Seeds exposed to TA
(3,000 ppm) and the positive control (PC) (10,000 ppm) did not germinate. (G) This visual perception
was confirmed by the quantitative determination of germination (%) of C. sinensis seeds under different
treatments. Six seeds were used in each treatment.

germination time compared to that of the negative control (NC), as evidenced by the
changes in seed color and size (Fig. 1). Seeds treated with the highest concentration of
TA (3,000 ppm) showed a stronger phytotoxic effect, and no formation of radicle was
observed (Fig. 1).
Following the observation of the phytotoxic activity of TA against the C. sinensis
seeds, we compared the metabolite profiles of the seeds exposed to the different treat-
ments by ultra-high-pressure liquid chromatography-mass spectrometry (UHPLC-MS).
Principal-component analysis (PCA) of quality control (QC), negative control (NC), her-
bicide (PC), and tryptoquialanine A (TA) extracts was performed to observe data repro-
ducibility and grouping tendencies (Fig. 2A). Data reproducibility was verified, as QC
samples formed a distinct cluster. The two principal components, PC1 and PC2, were
responsible for 37.9% of the variance of the data, revealing a separation between the
seed groups related to the treatment received (water, glyphosate herbicide, or TA).

January/February 2021 Volume 12 Issue 1 e03393-20 mbio.asm.org 3

 55
®
Costa et al.

Downloaded from http://mbio.asm.org/ on February 9, 2021 by guest

FIG 2 (A and B) PCA (A) and PLS-DA (B) for extracts of C. sinensis seeds treated with water (negative control), glyphosate
(herbicide), or tryptoquialanine A (TA).

In order to verify the classification of the seeds according to the treatment received,
partial least-squares discriminant analysis (PLS-DA) was performed. The PLS-DA score
plot confirmed a clear separation between the seed groups (Fig. 2B). In PLS-DA, PC1
and PC2 accounted for 26.9% of the variance (15.2% for PC1 and 11.7% for PC2). As in
the PCA score plot, control seeds were distributed in the opposite way of seeds treated
with TA along PC1, while seeds treated with herbicide were plotted in the center.
P. digitatum produces EVs in vitro and in vivo. To inhibit the germination of C.
sinensis seeds, TA is required to reach the extracellular environment. We then asked if
the extracellular export of TA could be vesicle-mediated. However, the production of
EVs by P. digitatum has not been reported so far. To address this question, we used
methods for EV detection in different models of P. digitatum growth. Specifically, the
production of EVs was evaluated in both solid agar medium and infected citrus fruits.
Transmission electron microscopy (TEM) of P. digitatum samples grown in vitro revealed
membranous structures with the typical features of vesicles, including round-shaped
structures with bilayered membranes in the 100- nm size range (Fig. 3A to D). Similar
results were observed for vesicles isolated from infected fruits. These results were con-
firmed by a second experimental approach. Nanoparticle tracking analysis (NTA) of the
same samples revealed particles mostly concentrated in the 100- to 200-nm range, with
subpopulations in the 200- to 300-nm and 300- to 400-nm size ranges (Fig. 3E and F). In
vitro and in vivo samples had similar properties, which were consistent with those previ-
ously described for fungal EVs (19, 25, 26).
Tryptoquialanine A is a component of P. digitatum EVs. The metabolite composi-
tion of the P. digitatum EVs was investigated by UHPLC-MS/MS in EV extracts obtained
in vivo (Fig. S2 and S3), followed by molecular networking in the Global Natural
Products Social Molecular Networking (GNPS) platform and, when available, compared
with standard metabolites. Molecular networking revealed three clusters (A, B, and C)
that exhibited compounds present in the EVs (Fig. 4, pink symbols). Metabolites were
manually identified by accurate mass analysis, MS/MS fragmentation profiles, or com-
parison with authentic standards (tryptoquialanine A and B) or as a hit in the GNPS data-
base. The observed signals corresponded, respectively, to tryptoquialanine A (m/z
519.19), tryptoquialanine B (m/z 505.17), deoxytryptoquialanine (m/z 503.19), cyclo-(Phe-
Val-Val-Tyr) (m/z 509.27), Phe-Val-Val-Phe (m/z 511.29), Phe-Val-Val-Tyr (m/z 527.28), and
cyclo-(Phe-Phe-Val-Val) (m/z 493.28).
In the molecular networking analysis, each consensus MS/MS spectrum is

January/February 2021 Volume 12 Issue 1 e03393-20 mbio.asm.org 4

 56
®
Extracellular Vesicles produced by P. digitatum

Downloaded from http://mbio.asm.org/ on February 9, 2021 by guest

FIG 3 Production of EVs by P. digitatum. (A to D) Analysis of ultracentrifugation pellets by TEM revealed the presence of the
typical round-shaped structures presenting double membranes. Similar results were obtained with samples obtained in vivo
(panels A and B) and in vitro (panels C and D). The visual observations using TEM were confirmed using NTA, which detected
particles mostly concentrated in the 100 to 200 nm range, with subpopulations in the 200 to 300 and 300 to 400 nm size ranges.
(E and F) Similar results were obtained with in vitro (E) and in vivo (F) samples. One representative experiment of three
independent replicates producing similar results is illustrated.

represented by a node, and all nodes are labeled with their precursor mass. Indole alka-
loids produced by P. digitatum were grouped in clusters A and B since they showed simi-
lar fragmentation patterns, with typical indole alkaloid fragments observed at [M1H]1
m/z 156.07, m/z 197.10, and m/z 213.10 (Fig. S4). Tryptoquialanines A and B and deoxy-
tryptoquialanine are the final products of the tryptoquialanine biosynthetic pathway
(27), and as already mentioned in this section, these indole alkaloids were reported as
major secondary metabolites for P. digitatum (16). In cluster C, the GNPS database indi-
cated the presence of cyclo-(Phe-Phe-Val-Val) (Fig. S5A), a mycotoxin known as fungis-
porin. We also observed that fungisporin analogues were grouped in this cluster. A frag-
mentation pattern with typical ions observed at [M1H]1 m/z 120.08, m/z 219.15, and m/
z 247.14 was previously described for compounds Phe-Val-Val-Phe and Phe-Val-Val-Tyr
(18, 28, 29) (Fig. S5B and C).
Quantification of tryptoquialanine A in P. digitatum EVs. The quantitative com-
position of alkaloids in P. digitatum EVs was evaluated using UHPLC-MS/MS analyses.
First, a calibration curve was prepared using standard TA (tR = 7.2 min) (Fig. S1) purified
from P. digitatum’s crude extracts (17). The coefficient of determination (r2) obtained
was greater than 0.998, indicating an excellent linearity (Fig. S6). Extracts of P. digitatum

January/February 2021 Volume 12 Issue 1 e03393-20 mbio.asm.org 5

 58
®
Extracellular Vesicles produced by P. digitatum

Downloaded from http://mbio.asm.org/ on February 9, 2021 by guest

FIG 5 P. digitatum EVs affect C. sinensis seeds. (A) Negative control (NC), consisting of C. sinensis
seeds incubated in PBS. (B) Positive control (PC), consisting of the citrus seeds incubated in glyphosate
(10,000 ppm). (C) Incubation of C. sinensis seeds with P. digitatum EVs produced in vivo (2.1  1010 EVs
ml21). Seeds exposed to fungal vesicles presented injured tissues (orange spots on their surface; red
arrows).

compounds present only in the infected fruits (blue, green, and yellow nodes) and
absent in control fruits (orange and pink nodes) (Fig. 6 and 7). Metabolites were man-
ually identified by their accurate masses and fragmentation profiles or identified as
hits in the GNPS database. Fragmentation patterns obtained by MS/MS analyses are
represented in Fig. S7. Accurate mass measurements showed mass errors below 5 ppm
(Table 1). The structures of the detected metabolites are shown in Fig. 8.
The compounds identified in our analysis included 15-dimethyl-2-epi-fumiquinazo-
line (m/z 460.19) A, deoxytryptoquialanone (m/z 459.16), tryptoquivaline L (m/z
433.15), tryptoquivaline Q (m/z 435.16 and 417.15), fumiquinazoline A (m/z 446.18 and
428.16), and fumiquinazoline C (m/z 444.16 and 426.15). Compounds 15-dimethyl-2-
epi-fumiquinazoline A and deoxytryptoquialanone are intermediates of the tryptoquia-
lanine biosynthetic pathway (27), while the tryptoquivalines and fumiquinazolines
were previously identified as P. digitatum metabolites (17). Fungisporin and analogues
were also identified in EVs (cluster C, Fig. 4). A few differences were observed in clus-
ters D, E, and F (Fig. 7 and 8) considering the production of secondary metabolites by
P. digitatum in different fruits. All identified compounds were detected in infected
plums (at 10 and 13 days postinoculation[dpi]) and oranges (at 7 dpi) (Fig. S8).
The complete molecular networking obtained for P. digitatum is represented in the
supplemental information (Fig. S9). P. digitatum molecular networking was composed
of 235 clusters, 83 of which (36%) were composed of unknown metabolites that were
only present in the infected fruits and absent in the control fruits. Molecular network-
ing also showed clusters containing unknown metabolites present only in infected
oranges or only in infected plums (Fig. S10).

DISCUSSION
Tryptoquialanines are the major secondary metabolites produced by P. digitatum
(16). The involvement of tryptoquialanines during the infection of citrus fruits by P.

January/February 2021 Volume 12 Issue 1 e03393-20 mbio.asm.org 7

 62
®
Extracellular Vesicles produced by P. digitatum

previously described for fungal proteins, glycans, and RNA (25, 33). In our model, EVs
were detected in culture and infected citrus fruits. The possibility of coisolation of plant
EVs in the in vivo samples cannot be ruled out, since it is well known that plant cells
also produce EVs during interaction with fungi (34). However, the similar features of
EVs obtained in vivo and in vitro and our vesicle compositional analysis reinforce the
notion that P. digitatum produces EVs in vitro and during plant infection. The observa-
tion of P. digitatum EVs gains additional significance considering that most of the stud-
ies characterizing fungal EVs used human pathogens as models, which implies that the
importance of EV production by phytopathogens has been underscored so far. In this
context, it has been only recently demonstrated that EVs from the cotton pathogen
Fusarium oxysporum f. sp. vasinfectum induce a phytotoxic response in plants (25). In
the EV cargo of F. oxysporum f. sp. vasinfectum, 482 enzymes were identified, including

Downloaded from http://mbio.asm.org/ on February 9, 2021 by guest

two polyketide synthases, yet the isolated EVs presented a deep purple color, indicat-
ing that a naphthoquinone pigment is packaged into the EVs (25). The authors sug-
gested that EVs could be a site of biosynthesis and transport of pigments and other
secondary metabolites (25), an idea that is quite complementary to what is presented
in our study.
Secondary metabolites participate in the virulence mechanisms of some phytopa-
thogenic fungi, implying that knowledge of metabolite exportation could improve the
understanding of the molecular basis of plant infection and fruit protection (4, 15).
However, the association of fungal metabolites and EVs has not been established so
far in plant infection models. Based on the observation of P. digitatum EVs in vitro
(potato dextrose agar) and in vivo (citrus fruits), we identified indole alkaloids and
mycotoxins in EV samples. Penicillium species are known to produce mycotoxins such
as fungisporin (35, 36). Fungisporin and analogues were reported in the cultures of P.
canescens (28), P. roqueforti (29), P. citrinum (18) and P. chrysogenum (37). Therefore,
the production of fungisporin compounds by P. digitatum was expected. To the best of
our knowledge, the presence of secondary metabolites and mycotoxins in EVs pro-
duced by a phytopathogen in vivo is reported here for the first time.
An estimate of the tryptoquialanine levels in EVs could provide insights into the
biosynthesis and metabolic flow of these molecules in P. digitatum. Previous studies
with oranges infected by P. digitatum showed that, at 5 days postinfection, TA was
detected in the orange epicarp, mesocarp, and endocarp, with concentrations of
24.810, 388, and 24 m g kg21, respectively (38). TA concentration in the EVs was consid-
erably lower. We then speculated that the biosynthesis of tryptoquialanines may occur
in the fungal cells with further export in EVs. This mechanism would differ from that
described by Bleackley et al. in the F. oxysporum f. sp. vasinfectum EVs (25).
P. digitatum EVs induced alterations in the C. sinensis seeds, as concluded from the
observation of color alteration resulting from tissue lesions. Similar results were
reported in recent studies with cotton cotyledons infiltrated with F. oxysporum f. sp.
vasinfectum. In this model, EVs induced discoloration around the sites of infiltration
(25). Therefore, the phytotoxic effect observed for the isolated TA was different from
that caused by the EVs. These differences were, in fact, expected, considering that ve-
sicular TA is accompanied by hundreds of other molecules. Those molecules could, for
instance, physically interact with TA, altering its relative concentration. In addition, if
those additional vesicular molecules have biological effects that differ from those
observed for TA alone, it would be very hard to predict what kind of effect would pre-
vail, since the relative concentration of vesicular molecules in the P. digitatum model is
still unknown. In any case, our results provide a proof-of-concept model showing that
P. digitatum exports bioactive molecules in EVs that can directly impact the pathogenic
process.
P. digitatum pathogenesis was believed to be restricted to citrus fruits (13).
However, this fungus is also an aggressive pathogen of stone fruits, including nectar-
ines and plums (39, 40). Few studies have investigated the infection of stone fruits by
P. digitatum. Even though P. digitatum disease was characterized at the physical

January/February 2021 Volume 12 Issue 1 e03393-20 mbio.asm.org 11

 63
®
Costa et al.

(incidence, lesion diameter, pH) and molecular (gene expression) levels (40, 41), no in-
formation on secondary metabolite production has been presented in the literature for
this host-pathogen interaction. Since tryptoquialanines and mycotoxins were found in
EVs produced during infection, the metabolic profile of P. digitatum in different fruits
was evaluated in order to verify if the same metabolites were found in the different
hosts. Molecular networking analyses indicate that intermediates of the tryptoquiala-
nine biosynthetic pathway are present in fruits and absent in EVs. These data are in
agreement with the quantification level of TA in EVs, reinforcing the idea that trypto-
quialanines are only transported by the EVs. Also, our results are the first to identify
the production of tryptoquialanines and other indole alkaloids in the P. digitatum-
stone fruit interaction. Likewise, the similarity between the metabolic profile in the
fruits suggests that the production of EVs by P. digitatum is not restricted to the citrus

Downloaded from http://mbio.asm.org/ on February 9, 2021 by guest

fruits, since the same metabolites found in EV cargo obtained from infection in citrus
were detected in plums. The clusters containing unknown metabolites present only in
infected oranges or only in infected plums (Fig. S9) suggest that the metabolite pro-
duction of P. digitatum can vary depending on the infected fruit.
Conclusions. This work is the first to report that P. digitatum is able to release EVs
and to report secondary metabolites in EVs produced by a phytopathogen in vivo.
Furthermore, we suggested that TA is synthesized intracellularly and exported in EVs.
Molecular networking confirmed our hypothesis that tryptoquialanines and mycotox-
ins are delivered through EVs during the infection process, since the intermediates of
the tryptoquialanine biosynthetic pathway are absent in the EVs. This delivery system
is not restricted to citrus and occurs in different types of fruits, such as plums.
A novel phytotoxic function for P. digitatum EVs and for tryptoquialanines was
observed. EVs caused alterations in the physiology of C. sinensis seed tissues, while TA
inhibited 100% of seed germination. The presence of alkaloids and mycotoxins in phy-
totoxic EVs opens new venues for the investigation of fungal secretion and its relation-
ship with plant pathogenesis. Also, our results provided new insights into the biologi-
cal role of the indole alkaloids and the infection strategies used by the phytopathogen
P. digitatum.

MATERIALS AND METHODS

Fungal strain and culture conditions. The P. digitatum strain is deposited in the Spanish Type
Culture Collection (CECT) (accession code CECT20796). The fungus was cultured in commercial potato
dextrose agar (PDA) (darkness, 7 days at 25°C). Conidial suspensions were prepared in sterile distilled
water and adjusted to a final concentration of 1.0  106 conidia ml21.
Purification of tryptoquialanine A by high-performance liquid chromatography (HPLC). P. digi-
tatum was cultivated in 12 liters of PDA distributed in petri dishes. After cultivation, the content of the
petri dishes was sliced and transferred to Erlenmeyer flasks. The content of the Erlenmeyer flasks was
extracted twice with ethyl acetate (EtOAc) under sonication in an ultrasonic bath for 1 h. The mixture of
agar, mycelia, and EtOAc was filtered, and the solvent was removed under reduced pressure.
The P. digitatum EtOAc extract was suspended in methanol (MeOH), filtered, and subjected to sepa-
ration by high-performance liquid chromatography (HPLC) in order to obtain pure tryptoquialanine A.
HPLC separation was performed with a Phenomenex column Luna 5-m m phenyl-hexyl (250  4.6 mm)
using a Shimadzu prominence HPLC LC-20AT instrument connected to a CBM-20A communication bus
module, to an SPD-M20A photodiode array detector, and to a SIL-20A auto sampler. The mobile phases
were 0.1% (vol/vol) formic acid in water (A) and acetonitrile (B). The flow rate was 1.0 ml min21. Elution
was performed as follows (A:B): gradient from 95:5 up to 55:45 for 30 min, then up to 35:65 from 30 to
52 min, then up to 5:95 from 52 to 55 min, remaining under this condition for 5 min. Column recondi-
tioning between each injection was a gradient to 95:5 from 60 to 61 min, remaining under this condition
for 9 min. Semipreparative HPLC separations were performed with a Phenomenex column Luna 5 m m
phenyl-hexyl (250  10 mm) using a Waters 1525 binary HPLC pump equipped with a Waters 2998 pho-
todiode array detector and a Waters fraction collector III. The eluent was the same as indicated above
with a flow rate of 4.7 ml min21.
Seed germination test (phytotoxicity assay). The phytotoxicity of tryptoquialanine A on seed ger-
mination was evaluated as previously described with a few modifications (42–45). Briefly, C. sinensis
seeds were manually collected from oranges purchased at a local grocery store (Campinas, São Paulo,
Brazil). Seeds coats were removed, and seeds were immersed in a 50% (vol/vol) commercial bleach solu-
tion for 15 min for surface sterilization. Six sterilized seeds were placed in each petri dish (6 cm) lined
with two filter papers. A volume of 2.5 ml of treatment solution was added to the plate. As the negative
control (NC), seeds were treated with sterile distilled H2O containing dimethyl sulfoxide (DMSO) 3% (vol/
vol). Tryptoquialanine A (TA) was solubilized in DMSO and diluted in sterile distilled water to a final

January/February 2021 Volume 12 Issue 1 e03393-20 mbio.asm.org 12

 64
®
Extracellular Vesicles produced by P. digitatum

concentration of 500, 1,000, and 3,000 ppm. The commercial herbicide Roundup was utilized as a posi-
tive control (PC) diluted to the concentrations of 10,000 and 3,000 ppm in sterile distilled water contain-
ing DMSO 3% (vol/vol). Treatment solutions were filtered through 0.22-m m membranes. Petri dishes
were sealed with tape and incubated in a biochemical oxygen demand (BOD) chamber at 25°C with
photoperiods of 12 h for 10 days. After incubation, the percentage of seed germination was calculated
as described in Equation 1, considering complete, proportionate, and healthy development.

number of germinated seeds

% Germination ¼  100 (1)
total number of seeds
To evaluate the phytotoxic activity of EVs, uncoated and sterilized C. sinensis seeds were placed in a
24-well cell culture plate lined with filter papers (1 seed per well). The seeds were treated with 100 m l of
a phosphate-buffered saline (PBS) solution of P. digitatum EVs (2.1  1010 EVs ml21). Negative controls
(NC) were performed using 100 m l of PBS, and for positive controls (PC), 100 m l of the herbicide Roundup
diluted in PBS (10,000 ppm) was used. The plate was sealed and incubated as described above.

Downloaded from http://mbio.asm.org/ on February 9, 2021 by guest

Infection of fruits by P. digitatum (in vivo assays) and metabolite extraction. For in vivo assays,
mature oranges (C. sinensis) and plums (Prunus salicina) obtained from a local grocery store (Campinas,
São Paulo, Brazil) were surfaced sterilized and wounded (17). Four fruits (2 oranges and 2 plums) were
infected with 15 m l of a P. digitatum 1.0  106 conidia ml21 solution. Control fruits (2 oranges and 2
plums) were also included. Infected and control fruits were stored in sterile 500-ml beakers in darkness
at 25°C. The fruits were incubated for different numbers of days postinoculation (dpi) in triplicates.
After the infection period (7 dpi for oranges, 10 and 13 dpi for plums), extraction of infected fruits
was performed as previously described, with few modifications (46). Fruits were cut around the infected
area (4 cm by 4 cm), and collected fruit pieces were extracted with 5 ml of MeOH for 1 h in ultrasonic
bath. The same procedure was performed for control fruits. MeOH extracts were filtered, dried with a N2
flux, and stored at 220°C.
Isolation of P. digitatum EVs and metabolite extraction. Isolation of P. digitatum EVs produced in
vitro was performed as previously described, with a few modifications (26). Fungal cells were cultivated
and softly scraped from PDA plates (triplicates, 20 ml of PDA per plate) using a sterile spatula. Fungal
mycelia were transferred to a Falcon tube filled with 30 ml of sterile phosphate-buffered saline (PBS). For
the analysis of EVs in vivo, nine oranges (C. sinensis) were infected with P. digitatum (as described above).
Infected fruits were incubated for 7 days (darkness, 25°C). Then, fungal cells in the infected areas of fruits
were softly scraped using a sterile spatula and transferred to a Falcon tube filled with 30 ml of PBS. Then,
30-ml cell suspensions obtained in vivo or in vitro were sequentially centrifuged to remove fungal cells
(5,000  g for 15 min at 4°C) and possible debris (15,000  g for 15 min at 4°C). The remaining superna-
tants were filtered through 0.45-m m-pore syringe filters and ultracentrifuged to collect EVs (100,000  g
for 1 h at 4°C). Ultracentrifugation pellets were negatively stained and analyzed by transmission electron
microscopy (TEM) as previously described (26). Briefly, EV samples were transferred to carbon- and
Formvar-coated grids and negatively stained with 1 % (vol/vol) uranyl acetate for 10 min. The grids were
then blotted dry before immediately being observed in a JEOL 1400Plus transmission electron micro-
scope at 90 kV. The same samples were subjected to nanoparticle tracking analysis (NTA) on an LM10
nanoparticle analysis system, coupled with a 488-nm laser and equipped with an SCMOS camera and a
syringe pump (Malvern Panalytical, Malvern, United Kingdom). Recorded data were acquired and ana-
lyzed using the NTA v.3.0 software (Malvern Panalytical).
To study the vesicular cargo, EVs obtained in vivo were extracted with 1 ml of MeOH HPLC grade for
1 h in an ultrasonic bath.
Mass spectrometry (MS) analyses. In vivo extracts. in vivo extracts were resuspended in 1 ml of
MeOH HPLC grade. An aliquot of 100 m l was diluted in 900 m l of MeOH HPLC grade, filtered through
0.22-m m membranes, and collected in glass vials. UHPLC-MS analyses were performed in a Waters
Acquity UPLC H-class chromatograph coupled to a Waters Xevo G2-XS QToF mass spectrometer using
electrospray ionization. The conditions were as follows: positive mode, capillary voltage at 1.2 kV; source
temperature at 100°C; cone gas (N2) flow of 50 liters h21; desolvation gas (N2) flow of 750 liters h21, and
m/z range of 100 to 1,500. MS/MS analyses were performed using a collision energy ramp of 6 to 9 V
(low mass) and 60 to 80 V (high mass). A BEH C18 column (2.1 mm by 100 mm by 1.7 m m) was used.
Mobile phases were 0.1% (vol/vol) formic acid in water (A) and acetonitrile (B). Eluent profile (A:B) 0 to
6 min, gradient from 90:10 up to 50:50; 6 to 9 min, gradient up to 2:98; 9 to 10 min, gradient up to 90:10.
The flow rate was 0.2 ml min21. The injection volume was 2 m l. Operation and spectrum analyses were
conducted using Waters MassLynx v.4.1. software.
P. digitatum EV extracts. First, 1 ml of EV extracts was filtered through 0.22-m m membranes into
glass vials. UHPLC-MS analyses were performed using a Thermo Scientific QExactive hybrid Quadrupole-
Orbitrap mass spectrometer with the following parameters: electrospray ionization in positive mode,
capillary voltage at 13.5 kV; capillary temperature at 250°C; S-lens of 50 V, and m/z range of 133.40 to
2,000.00. MS/MS was performed using normalized collision energy (NCE) of 30 eV, and 5 precursors per
cycle were selected. Stationary phase: Thermo Scientific Accucore C18 2.6 m m (2.1 mm x 100 mm) col-
umn. Mobile phases were 0.1% (vol/vol) formic acid in water (A) and acetonitrile (B). Eluent profile (A:B)
0 to 10 min, gradient from 95:5 up to 2:98; held for 5 min; 15 to 16.2 min gradient up to 95:5; held for
8.8 min. The flow rate was 0.2 ml min21. The injection volume was 3 m l. Operation and spectrum analyses
were conducted using Xcalibur software (v.3.0.63) developed by Thermo Fisher Scientific.
Seed extracts. Two seeds of each condition, TA (3,000 ppm), PC (10,000 ppm), and NC, were macer-
ated with liquid nitrogen in triplicate. Aliquots of 100 mg of macerated seeds were extracted in plastic

January/February 2021 Volume 12 Issue 1 e03393-20 mbio.asm.org 13

 65
®
Costa et al.

tubes with 2 ml of MeOH containing 0.1% (vol/vol) formic acid during 1 h in an ultrasonic bath. The
extracts were filtered (0.22 m m), dried with a N2 flux, and stored at 220°C.
Seed extracts were resuspended in 1 ml of MeOH and aliquots of 100 m l and diluted with 900 m l and
filtered through a 0.22-m m membrane. UHPLC-MS analyses were performed using a Thermo Scientific
QExactive hybrid Quadrupole-Orbitrap mass spectrometer with the following parameters: electrospray
ionization in positive mode, capillary voltage at 3.5 kV; capillary temperature at 300°C; S-lens of 50 V,
and m/z range of 100.00 to 1,500.00. MS/MS was performed using normalized collision energy (NCE) of
20, 30, and 40 eV, and a maximum of 5 precursors per cycle were selected. A Waters Acquity UPLC BEH
C18 1.7-m m (2.1 mm by 50 mm) column was used. Mobile phases were 0.1% (vol/vol) formic acid in
water (A) and acetonitrile (B). Eluent profile (A:B) 0 to 10 min, gradient from 95:5 up to 2:98; held for
5 min; 15 to 16.2 min gradient up to 95:5; held for 3.8 min. The flow rate was 0.2 ml min21. UHPLC-MS
operation and spectrum analyses were performed using Xcalibur software (v.3.0.63). Samples were
injected in random order. A quality control (QC) sample was prepared with 50 m l of each sample and
was injected three times at the beginning of the batch and after three sample injections (47–49).
Quantification of tryptoquialanine A. Standard TA isolated from P. digitatum and EV extract were

Downloaded from http://mbio.asm.org/ on February 9, 2021 by guest

analyzed using a Waters Acquity UPLC system coupled to a Waters Micromass Quattro Micro TM API
with electrospray ionization source and a triple quadrupole mass analyzer. Analyses were performed in
the positive mode with an m/z range of 100 to 1,200, capillary voltage of 3 kV, cone voltage of 25 V, inlet
capillary temperature of 150°C, and nebulizing gas temperature of 200°C. Stationary phase: Thermo
Scientific column Accucore C18 2.6 m m (2.1 mm by 100 mm). Mobile phase: 0.1% formic acid (A) and ace-
tonitrile (B). Eluent profile (A/B): 95/5 up to 2/98 within 10 min, held for 5 min, up to 95/5 within 1.2 min,
and held for 3.8 min. The total run time was 20 min for each run, and the flow rate was 0.2 ml min21.
Injection volume: 10 m l. All the operation and spectrum analyses were conducted using Waters
MassLynx v.4.1.
For the construction of the calibration curve, standard TA was diluted in the range concentration of
6.25 to 0.006 m g ml21, and selected reaction monitoring (SRM) analyses were performed following con-
ditions as previously described: m/z 519 ! 197 (quantification) and m/z 519 ! 213 (monitoring), colli-
sion energy of 22 eV (38).
For quantification of tryptoquialanine A in EVs, 40 m l of a 2.1  1010 EV ml21 solution was dried and
extracted with 100 m l of MeOH HPLC grade as previously described. Then, 100 m l of EV extract solution
was transferred to glass vials and analyses were performed in duplicate.
Statistical and metabolomic analyses. Feature detection was performed on XCMS online (v.3.5.1)
using the following parameters: method: centWave, prefilter peaks and intensity: 3 and 5,000, ppm: 2.5,
Signal/noise threshold: 10, peak width: 5 to 20, mzdiff: 0.01, and noise filter: 1,000. Preprocessing
included median fold change normalization on XCMS Online. Multivariate and univariate analyses of the
feature list were performed with the MetaboAnalyst tool (v.4.0). Pareto scaling was applied. One-way
analysis of variance (ANOVA) was performed, and all the results were analyzed using a confidence level
of 95% and a significance level corresponding to P , 0.05. Principal-component analysis (PCA) was per-
formed for an exploratory analysis, followed by partial least-squares discriminant analysis (PLS-DA). A
permutation test (cross validation) was performed to determine the reliability of the created PLS-DA
model.
Molecular networking analyses. MS data were converted to mzXML format using MSConvert GUI,
a tool of the ProteoWizard package. Molecular networks for in vivo assays and EV extracts were created
using the mzXML files on the online workflow at the Global Natural Products Social Molecular
Networking (GNPS) platform (http://gnps.ucsd.edu). Data were filtered by removing all MS/MS peaks
within 617 Da of the precursor ion. MS/MS spectra were window filtered by choosing only the top 6
peaks in the 650-Da window throughout the spectrum. The data were then clustered with MS-Cluster
with a parent mass tolerance of 0.02 Da and an MS/MS fragment ion tolerance of 0.02 Da to create con-
sensus spectra. Consensus spectra that contained fewer than 2 spectra were discarded. A network was
then created where edges were filtered to have a cosine score above 0.6 and more than 5 matched
peaks. The spectra in the network were then searched against GNPS’s spectral libraries. The library spec-
tra were filtered in the same manner as the input data. All matches between network spectra and library
spectra were required to have a score above 0.6 and at least 5 matched peaks (50).

SUPPLEMENTAL MATERIAL
Supplemental material is available online only.
FIG S1, TIF file, 0.2 MB.
FIG S2, TIF file, 0.2 MB.
FIG S3, TIF file, 0.3 MB.
FIG S4, TIF file, 0.1 MB.
FIG S5, TIF file, 0.1 MB.
FIG S6, TIF file, 0.01 MB.
FIG S7, TIF file, 1.1 MB.
FIG S8, TIF file, 0.3 MB.
FIG S9, JPG file, 2.6 MB.
FIG S10, TIF file, 0.03 MB.

January/February 2021 Volume 12 Issue 1 e03393-20 mbio.asm.org 14


Extracellular Vesicles produced by P. digitatum
ACKNOWLEDGMENTS
This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal
de Nível Superior - Brasil (CAPES) - Finance Code 001, Fundação de Amparo a Pesquisa
no Estado de São Paulo (grant numbers 2019/11563-2, 2019/06359-7, 2017/24462-4,
2013/50228-8, and 2015/01017-0) L’Oréal Brazil, together with ABC and UNESCO in
Brazil. M.L.R. was supported by grants from the Brazilian Ministry of Health (grant
440015/2018-9), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq;
grants 405520/2018-2 and 301304/2017-3), and Fiocruz (grants PROEP-ICC 442186/2019-
3, VPPCB-007-FIO-18, and VPPIS-001-FIO18). We also acknowledge support from the
Instituto Nacional de Ciência e Tecnologia de Inovação em Doenças de Populações
Negligenciadas (INCT-IDPN).
Conflict of interest: M.L.R. is currently on leave from the position of associate
professor at the Microbiology Institute of the Federal University of Rio de Janeiro, Brazil.
REFERENCES
1. Jr-Mattos D, Carlos EF. 2018. The role of the International Society of Citri-
culture on the world citrus industry. Citrus Res Technol 38:228–232.
https://doi.org/10.4322/crt.ICC171.
2. Citrus: World Markets and Trade. 2020. United States Department of Agri-
culture. Foreign Agricultural Service. https://downloads.usda.library
.cornell.edu/usda-esmis/files/w66343603/00000g55g/kp78h0193/citrus
.pdf. Accessed 14 April 2020.
3. Blauth de Lima F, Félix C, Osório N, Alves A, Vitorino R, Domingues P,
Correia A, Ribeiro RTS, Esteves AC. 2016. Secretome analysis of Tricho-
derma atroviride T17 biocontrol of Guignardia citricarpa. Biol Control
99:38–86. https://doi.org/10.1016/j.biocontrol.2016.04.009.
4. Costa JH, Bazioli JM, Pontes JGM, Fill TP. 2019. Penicillium digitatum infec-
tion mechanisms in citrus: what do we know so far? Fungal Biol
123:584–593. https://doi.org/10.1016/j.funbio.2019.05.004.
5. Qian X, Yang Q, Zhang Q, Abdelhai MH, Dhanasekaran S, Serwah BNA,
Gu N, Zhang H. 2019. Elucidation of the initial growth process and the
infection mechanism of Penicillium digitatum on postharvest citrus
(Citrus reticulata blanco). Microorganisms 7:485. https://doi.org/10
.3390/microorganisms7110485.
6. Chen J, Shen Y, Chen C, Wan C. 2019. Inhibition of key citrus postharvest
fungal strains by plant extracts in vitro and in vivo: a review. Plants 8:26.
https://doi.org/10.3390/plants8020026.
7. Zhang S, Zheng Q, Xu B, Liu J. 2019. Identification of the fungal pathogens
of postharvest disease on peach fruits and the control mechanisms of Bacil-
lus subtilis JK-14. Toxins 11:322. https://doi.org/10.3390/toxins11060322.
8. Spadaro D, Droby S. 2016. Development of biocontrol products for post-
harvest diseases of fruit: the importance of elucidating the mechanisms
of action of yeast antagonists. Trends Food Sci Technol 47:39–49. https://
doi.org/10.1016/j.tifs.2015.11.003.
9. Sanzani SM, Reverberi M, Geisen R. 2016. Mycotoxins in harvested fruits
and vegetables: insights in producing fungi, biological role, conducive
conditions, and tools to manage postharvest contamination. Postharvest
Biol Tec 122:95–105. https://doi.org/10.1016/j.postharvbio.2016.07.003.
10. Nicosia MGLD, Pangallo S, Raphael G, Romeo FV, Strano MC, Rapisarda P,
Droby S, Schena L. 2016. Control of postharvest fungal rots on citrus fruit
and sweet cherries using a pomegranate peel extract. Postharvest Biol
Tec 114:54–61. https://doi.org/10.1016/j.postharvbio.2015.11.012.
11. Dukare AS, Paul S, Nambi VE, Gupta RK, Singh R, Sharma K, Vishwakarma
RK. 2019. Exploitation of microbial antagonists for the control of post-
harvest diseases of fruits: a review. Crit Rev Food Sci Nutr 59:1498–1513.
https://doi.org/10.1080/10408398.2017.1417235.
12. Bazioli JM, Belinato JR, Costa JH, Akiyama DY, Pontes JGM, Kupper KC,
Augusto F, Carvalho JE, Fill TP. 2019. Biological control of citrus postharvest
phytopathogens. Toxins 11:460. https://doi.org/10.3390/toxins11080460.
13. Ramón-Carbonell M, Sánchez-Torres P. 2017. The transcription factor
PdSte12 contributes to Penicillium digitatum virulence during citrus
fruit infection. Postharvest Biol Tec 125:129–139. https://doi.org/10
.1016/j.postharvbio.2016.11.012.
14. Vu TX, Ngo TT, Mai LTD, Bui T, Le DH, Bui HTV, Nguyen HQ, Ngo BX, Tran
V. 2018. A highly efficient Agrobacterium tumefaciens-mediated transfor-
mation system for the postharvest pathogen Penicillium digitatum using
January/February 2021 Volume 12 Issue 1 e03393-20
66
®
Downloaded from http://mbio.asm.org/ on February 9, 2021 by guest
DsRed and GFP to visualize citrus host colonization. J Microbiol Methods
144:134–144. https://doi.org/10.1016/j.mimet.2017.11.019.
15. Scharf DH, Heinekamp T, Brakhage AA. 2014. Human and plant fungal
pathogens: the role of secondary metabolites. PLoS Pathog 10:e1003859.
https://doi.org/10.1371/journal.ppat.1003859.
16. Ariza MR, Larsen TO, Peterson BO, Duus JO, Barrero AF. 2002. Penicillium
digitatum metabolites on synthetic media and citrus fruits. J Agric Food
Chem 50:6361–6365. https://doi.org/10.1021/jf020398d.
17. Costa JH, Bazioli JM, Araújo EV, Vendramini PH, Porto MCF, Eberlin MN,
Souza-Neto JA, Fill TP. 2019. Monitoring indole alkaloid production by Peni-
cillium digitatum during infection process in citrus by mass spectrometry
imaging and molecular networking. Fungal Biol 123:594–600. https://doi
.org/10.1016/j.funbio.2019.03.002.
18. Costa JH, Wassano CI, Angolini CFF, Scherlach K, Hertweck C, Fill TP. 2019.
Antifungal potential of secondary metabolites involved in the interaction
between citrus pathogens. Sci Rep 9:18647. https://doi.org/10.1038/
s41598-019-55204-9.
19. Souza JAM, Baltazar LM, Carregal VM, Gouveia-Eufrasio L, Oliveira AG,
Dias WG, Rocha MC, Miranda KR, Malavazi I, Santos DA, Frézard FJG,
Souza DG, Teixeira MM, Soriani FM. 2019. Characterization of Aspergillus
fumigatus extracellular vesicles and their effects on macrophages and
neutrophils functions. Front Microbiol 10:2008. https://doi.org/10.3389/
fmicb.2019.02008.
20. Herkert PF, Amatuzzi RF, Alves LR, Rodrigues ML. 2019. Extracellular
vesicles as vehicles for the delivery of biologically active fungal mole-
cules. Curr Protein Pept Sci 20:1027–1036. https://doi.org/10.2174/
1389203720666190529124055.
21. Silva VKA, Rodrigues ML, May RC. 2019. Deciphering fungal extracellular
vesicles: from cell biology to pathogenesis. Curr Clin Micro Rep 6:89–97.
https://doi.org/10.1007/s40588-019-00128-1.
22. Rybak K, Robatzek S. 2019. Functions of extracellular vesicles in immunity
and virulence. Plant Physiol 179:1236–1247. https://doi.org/10.1104/pp
.18.01557.
23. Regente M, Pinedo M, Clemente HS, Balliau T, Jamet E, Canal L. 2017.
Plant extracellular vesicles are incorporated by a fungal pathogen and in-
hibit its growth. J Exp Bot 68:5485–5496. https://doi.org/10.1093/jxb/
erx355.
24. Bielska E, Sisquella MA, Aldeieg M, Birch C, O’Donoghue EJ, May RC. 2018.
Pathogen-derived extracellular vesicles mediate virulence in the fatal
human pathogen Cryptococcus gattii. Nat Commun 9:1556. https://doi
.org/10.1038/s41467-018-03991-6.
25. Bleackley MR, Samuel M, Garcia-Ceron D, McKenna JA, Lowe RGT, Pathan
M, Zhao K, Ang C-S, Mathivanan S, Anderson MA. 2019. Extracellular
vesicles from the cotton pathogen Fusarium oxysporum f. sp. vasinfectum
induce a phytotoxic response in plants. Front Plant Sci 10:1610. https://
doi.org/10.3389/fpls.2019.01610.
26. Reis FCG, Borges BS, Jozefowicz LJ, Sena BAG, Garcia AWA, Medeiros LC,
Martins ST, Honorato L, Schrank A, Vainstein MH, Kmetzsch L, Nimrichter
L, Alves LR, Staats CC, Rodrigues ML. 2019. A novel protocol for the isola-
tion of fungal extracellular vesicles reveals the participation of a putative
scramblase in polysaccharide export and capsule construction in
mbio.asm.org 15

Costa et al.
Cryptococcus gattii. mSphere 4:e00080-19. https://doi.org/10.1128/mSphere
.00080-19.
27. Gao X, Chooi YH, Ames BD, Wang P, Walsh CT, Tang Y. 2011. Fungal indole
alkaloid biosynthesis: genetic and biochemical investigation of the tryp-
toquialanine pathway in Penicillium aethiopicum. J Am Chem Soc
133:2729–2741. https://doi.org/10.1021/ja1101085.
28. Bertinetti BV, Peña NI, Cabrera GM. 2009. An antifungal tetrapeptide from
the culture of Penicillium canescens. Chem Biodivers 6:1178–1184. https://
doi.org/10.1002/cbdv.200800336.
29. Hammerl R, Frank O, Schmittnägel T, Ehrmann MA, Hofmann T. 2019.
Functional metabolome analysis of Penicillium roqueforti by means of dif-
ferential off-line LC-NMR. J Agric Food Chem 67:5135–5146. https://doi
.org/10.1021/acs.jafc.9b00388.
30. Zhu C, Sheng D, Wu X, Wang M, Hu X, Li H, Yu D. 2017. Identification of
secondary metabolite biosynthetic gene clusters associated with the
infection of citrus fruit by Penicillium digitatum. Postharvest Biol Tec
134:17–21. https://doi.org/10.1016/j.postharvbio.2017.07.011.
31. Wang XD, Sun C, Gao S, Wang L, Shuokui H. 2001. Validation of germina-
tion rate and root elongation as indicator to assess phytotoxicity with
Cucumis sativus. Chemosphere 44:1711–1721. https://doi.org/10.1016/
s0045-6535(00)00520-8.
32. Ashagre H, Almaw D, Feyisa T. 2013. Effect of copper and zinc on seed
germination, phytotoxicity, tolerance and seedling vigor of tomato (Lyco-
persicon esculentum L. cultivar ROMA VF). Int J Agric Sci Res 2:312–317.
33. Zamith-Miranda D, Nimrichter L, Rodrigues ML, Nosanchuk JD. 2018. Fun-
gal extracellular vesicles: modulating host-pathogen interactions by both
the fungus and the host. Microbes Infect 20:501–504. https://doi.org/10
.1016/j.micinf.2018.01.011.
34. Cai Q, He B, Weiberg A, Buck AH, Jin H. 2019. Small RNAs and extracellular
vesicles: new mechanisms of cross-species communication and innovate
tools for disease control. PLoS Pathog 15:e1008090. https://doi.org/10
.1371/journal.ppat.1008090.
35. Frisvad JC, Smedsgaard J, Larsen TO, Samson RA. 2004. Mycotoxins, drugs
and other extrolites produced by species in Penicillium subgenus Penicil-
lium. Stud Mycol 49:201–241.
36. Oppong-Danquah E, Passaretti C, Chianese O, Blümel M, Tasdemir D.
2020. Mining the metabolome and the agricultural and pharmaceutical
potential of sea foam-derived fungi. Mar Drugs 18:128. https://doi.org/10
.3390/md18020128.
37. Ali H, Ries MI, Lankhorst PP, Hoeven RAM, Schouten OL, Noga M,
Hankemeier T, Peij NNME, Bovenberg RAL, Vreeken RJ, Driessen AJM.
2014. A non-canonical NRPS is involved in the synthesis of fungisporin
and related hydrophobic cyclic tetrapeptides in Penicillium chrysogenum.
PLoS One 9:e98212. https://doi.org/10.1371/journal.pone.0098212.
38. Araújo EV, Vendramini PH, Costa JH, Eberlin MN, Montagner CC, Fill TP.
2019. Determination of tryptoquialanines A and C produced by Penicil-
lium digitatum in oranges: are we safe? Food Chem 301:125285. https://
doi.org/10.1016/j.foodchem.2019.125285.
39. Navarro D, Diáz-Mula HM, Guillén F, Zapata PJ, Castillo S, Serrano M,
Valero D, Martínez-Romero D. 2011. Reduction of nectarine decay caused
by Rhizopus stolonifer, Botrytis cinerea and Penicillium digitatum with Aloe
January/February 2021 Volume 12 Issue 1 e03393-20
67
®
vera gel alone or with the addition of thymol. Int J Food Microbiol
151:241–246. https://doi.org/10.1016/j.ijfoodmicro.2011.09.009.
40. Louw JP, Korsten L. 2016. Postharvest decay of nectarine and plum
caused by Penicillium spp. Eur J Plant Pathol 146:779–791. https://doi.org/
10.1007/s10658-016-0956-0.
41. Louw JP, Korsten L. 2019. Impact of ripeness on the infection and coloni-
sation of Penicillium digitatum and P expansum on plum. Postharvest Biol
Technol 149:148–158. https://doi.org/10.1016/j.postharvbio.2018.11.024.
42. Niedz RP. 2008. In vitro germination of citrus seed. Proc Fla State Hort Soc
121:148–151.
43. Habermann E, Pereira VDC, Imatomi M, Pontes FC, Gualtieri CJ. 2017. In
vitro herbicide activity of crude and fractionated leaf extracts of Blepharo-
calyx salicifolius (Myrtaceae). Braz J Bot 40:33–40. https://doi.org/10.1007/
s40415-016-0317-4.
44. Abdelgaleil SAM, Saad MMG, Ariefta NR, Shiono Y. 2020. Antimicrobial
and phytotoxic activities of secondary metabolites from Haplophyllum
Downloaded from http://mbio.asm.org/ on February 9, 2021 by guest
tuberculatum and Chrysanthemum coronarium. S Afr J Bot 128:35–41.
https://doi.org/10.1016/j.sajb.2019.10.005.
45. Gris D, Boaretto AG, Marques MR, Damasceno-Junior G, Carollo CA. 2019.
Secondary metabolites that could contribute to the monodominance of
Erythrina fusca in the Brazilian Pantanal. Ecotoxicology 28:1232–1240.
https://doi.org/10.1007/s10646-019-02133-y.
46. Smedsgaard J. 1997. Micro-scale extraction procedure for standardized
screening of fungal metabolite production in cultures. J Chromatogr A
760:264–270. https://doi.org/10.1016/S0021-9673(96)00803-5.
47. Li X, Zhang X, Ye L, Kang Z, Jia D, Yang L, Zhang B. 2019. LC-MS-based
metabolomic approach revealed the significantly different metabolic pro-
files of five commercial truffle species. Front Microbiol 10:2227. https://
doi.org/10.3389/fmicb.2019.02227.
48. Albóniga OE, González O, Alonso RM, Xu Y, Goodacre R. 2020. Optimiza-
tion of XCMS parameters for LC-MS metabolomics: an assessment of
automated versus manual tuning and its effect on the final results.
Metabolomics 16:14. https://doi.org/10.1007/s11306-020-1636-9.
49. Le TN, da Silva D, Colas C, Darrouzet E, Baril P, Leseurre L, Maunit B. 2020.
Development of an LC-MS multivariate nontargeted methodology for dif-
ferential analysis of the peptide profile of Asian hornet venom (Vespa
velutina nigrithorax): application to the investigation of the impact of col-
lection period variation. Anal Bioanal Chem 412:1419–1430. https://doi
.org/10.1007/s00216-019-02372-2.
50. Wang M, Carver JJ, Phelan VV, Sanchez LM, Garg N, Peng Y, Nguyen DD,
Watrous J, Kapono CA, Luzzatto-Knaan T, Porto C, Bouslimani A, Melnik
AV, Meehan MJ, Liu WT, Crüsemann M, Boudreau PD, Esquenazi E,
Sandoval-Calderón M, Kersten RD, Pace LA, Quinn RA, Duncan KR, Hsu
CC, Floros DJ, Gavilan RG, Kleigrewe K, Northen T, Dutton RJ, Parrot D,
Carlson EE, Aigle B, Michelsen CF, Jelsbak L, Sohlenkamp C, Pevzner P,
Edlund A, McLean J, Piel J, Murphy BT, Gerwick L, Liaw CC, Yang YL,
Humpf HU, Maansson M, Keyzers RA, Sims AC, Johnson AR, Sidebottom
AM, Sedio BE, et alet al. 2016. Sharing and community curation of mass
spectrometry data with Global Natural Products Social Molecular Net-
working. Nat Biotechnol 34:828–837. https://doi.org/10.1038/nbt.3597.
mbio.asm.org 16
 68

4.3 Informações suplementares do capítulo II

Phytotoxic tryptoquialanines produced in vivo by Penicillium digitatum are

exported in extracellular vesicles

Jonas Henrique Costaa, Jaqueline Moraes Baziolia,b, Luidy Darllan Barbosaa, Pedro
Luis Theodoro dos Santos Júniora, Flavia C. G. Reisc,d, Tabata Klimeckc, Camila
Manoel Crnkovice, Roberto G. S. Berlinckf, Alessandra Sussulinia, Marcio L.
Rodriguesc,g#, Taícia Pacheco Filla#

a Institute of Chemistry, University of Campinas, CP 6154, 13083-970, Campinas – SP,

Brazil
b Faculty of Pharmaceutical Sciences, University of Campinas, 13083-859, Campinas,
SP, Brazil
c Instituto Carlos Chagas, Fundação Oswaldo Cruz (Fiocruz), Curitiba – PR, Brazil,
d Centro de Desenvolvimento Tecnológico em Saúde (CDTS), Fiocruz, Rio de
Janeiro, Brazil)
e Department of Biochemical and Pharmaceutical Technology, School of
Pharmaceutical Sciences, University of São Paulo, 05508-000, São Paulo – SP, Brazil
f Instituto de Química de São Carlos, Universidade de São Paulo, CP 780, 13560-
970, São Carlos - SP, Brazil
g Instituto de Microbiologia Paulo de Góes (IMPG), Universidade Federal do Rio de
Janeiro, Rio de Janeiro - RJ, Brazil

#Corresponding authors: phone +55-19-35213092, e-mail taicia@unicamp.br;

marcio.rodrigues@fiocruz.br
 69

FIG S1 HPLC chromatograms (280 nm) for (A) P. digitatum extract and (B) purified tryptoquialanine A.
HPLC-MS analysis of purified tryptoquialanine A (m/z 519): (C) extracted ion chromatogram and (D)
mass spectrum.
70
71
 72

FIG S2 Extracted ion chromatograms of m/z 519.18 for (A) P. digitatum EVs extract, (B) MS analysis
control and (C) EV isolation methodology control. (D) Mass spectrum of ion [M+H] + m/z 519.1877
obtained for tryptoquialanine A (error = 0.61 ppm) at 9.2 min. Extracted ion chromatograms of m/z
505.17 for (E) P. digitatum EV extract, (F) MS analysis control and (G) EV isolation methodology control.
(H) Mass spectrum of ion [M+H]+ m/z 505.1719 obtained for tryptoquialanine B (error = 0.26 ppm) at 8.7
min. Extracted ion chromatograms of m/z 503.19 for (I) P. digitatum EV extract, (J) MS analysis control
and (K) EV isolation methodology control. (L) Mass spectrum of ion [M+H] + m/z 505.1926 obtained for
deoxytryptoquialanine (error = 0.23 ppm) at 8.6 min.
73
74
75
 76

FIG S3 Extracted ion chromatograms of m/z 509.27 for (A) P. digitatum EV extract, (B) MS analysis
control and (C) EV isolation methodology control. (D) Mass spectrum of ion [M+H] + m/z 509.2761
obtained for cyclo-(Phe-Val-Val-Tyr) (error = 0.48 ppm) at 8.6 min. Extracted ion chromatograms of m/z
511.29 for (E) P. digitatum EV extract, (F) MS analysis control and (G) EV isolation methodology control.
(H) Mass spectrum of ion [M+H]+ m/z 511.2917 obtained for Phe-Val-Val-Phe (error = 0.43 ppm) at 7.1
min. Extracted ion chromatograms of m/z 527.28 for (I) P. digitatum EV extract, (J) MS analysis control
and (K) EV isolation methodology control. (L) Mass spectrum of ion [M+H] + m/z 527.2866 obtained for
Phe-Val-Val-Tyr (error = 0.40 ppm) at 6.4 min. Extracted ion chromatograms of m/z 493.28 for (M) P.
digitatum EV extract, (N) MS analysis control and (O) EV isolation methodology control. (P) Mass
spectrum of ion [M+H]+ m/z 493.2809 obtained for cyclo-(Phe-Phe-Val-Val) (error = 0.03 ppm) at 9.7
min.
77
 78

FIG S4 Comparison of MS/MS spectra between (A) P. digitatum EV extract and (B) purified
tryptoquialanine A, (C) P. digitatum EV extract and (D) purified tryptoquialanine B, (E) P. digitatum EV
extract and (F) purified deoxytryptoquialanine.
 79

FIG S5 (A) MS/MS match between GNPS database (green) and compound cyclo-(Phe-Phe-Val-Val)
(black). MS/MS spectrum of (B) m/z 511.29 and (C) m/z 527.28 obtained from P. digitatum EV extract.
Fragmentation pattern was compared with MS/MS data of compounds Phe-Val-Val-Phe and Phe-Val-
Val-Tyr reported in the literature.
 80

FIG S6 UHPLC-MS/MS calibration curve obtained for tryptoquialanine A.

81
82
83
84
85
 86

FIG S7 Mass spectrum and MS/MS spectrum for (A) tryptoquialanine A ion [M+H] + m/z 519.1876, (B)
tryptoquialanine B ion [M+H]+ m/z 505.1735, (C) deoxytryptoquialanine ion [M+H]+ m/z 503.1948, (D)
 87

cyclo-(Phe-Val-Val-Tyr) ion [M+H]+ m/z 509.2770, (E) Phe-Val-Val-Phe ion [M+H]+ m/z 511.2920, (F)
Phe-Val-Val-Tyr ion [M+H]+ m/z 527.2859, (G) cyclo-(Phe-Phe-Val-Val) ion [M+H]+ m/z 493.2822, (H)
15-dimethyl-2-epi-fumiquinazoline A ion [M+H]+ m/z 460.1983, (I) deoxytryptoquialanone ion [M+H]+ m/z
459.1678, (J) tryptoquivaline L ion [M+H]+ m/z 433.1523, (K) tryptoquivaline Q ion [M+H]+ m/z 435.1673,
(L) fumiquinazoline A ion [M+H]+ m/z 446.1838 and (M) fumiquinazoline C ion [M+H]+ m/z 444.1687.

FIG S8 Fruits infected with P. digitatum. (A) plum at 10 dpi, (B) plum at 13 dpi and (C) orange at 7 dpi.
 88

FIG S9 Complete molecular networking obtained for extracts of plums and citrus fruits infected by P.
digitatum.
 89

FIG S10 Examples of clusters obtained by molecular networking for extracts of P. digitatum infection in
different fruits. The unknown compounds are present just in one type of the fruits and absent in the
control fruits
 90

5 CAPÍTULO III
Explorando a interação entre flavonóides cítricos e fungos fitopatógenicos
através de atividades enzimáticas

O conteúdo deste capítulo é composto pelo artigo intitulado “Exploring the

interaction between citrus flavonoids and phytopatogenic fungi through
enzymatic activities”, publicado no periódico Bioorganic Chemistry. A reprodução
deste documento pelos autores, para fins não comerciais, é autorizada pela
editora Elsevier e não necessita de permissão escrita (anexo 10.1).

Referência: Costa, J.H., Fernandes, L.S., Akiyama, D.Y., et al. 2020. Exploring
the interaction between citrus flavonoids and phytopatogenic fungi through
enzymatic activities. Bioorganic Chemistry, v. 102, 104126.
doi: 10.1016/j.bioorg.2020.104126

Versão final publicada disponível em:

https://www.sciencedirect.com/science/article/pii/S0045206820314231

5.1 Resumo
Neste estudo, ensaios de biocatalíse in vitro foram aplicados utilizando
flavonoides (hesperidina e naringina) e os principais fungos patógenos de frutas
cítricas. O objetivo era explorar as atividades enzimáticas envolvidas nesta interação,
uma vez que os flavonoides estão envolvidos na defesa dos citros contra
fitopatógenos. Através de análises de LC-MS observou-se que os fitopatógenos
hidrolisam os flavonoides através da atividade enzimática de naringinases e
hesperidinases. Em 7 dias, P. digitatum e Penicillium italicum exibiram as maiores
taxas de hidrólise sob os flavonoides, atingindo conversões maior que 90 %. Ainda,
Geothrichum citri-aurantii não apresentou atividade enzimática sob nenhum dos
flavonóides e Penicillium expansum apenas foi capaz de hidrolisar a hesperidina. Para
avaliar a interaçao enzimática entre fitopatógenos e flavonóides in vivo, laranjas
infectadas com P. digitatum foram analisadas por IMS e redes moleculares. Os
ensaios in vivo permitiram detectar novo flavonoides hidroxilados pela planta em
resposta a infecção. A relação entre os flavonoides e as atividades enzimáticas
reportadas podem auxiliar no entendimento da interação entre patógeno-hospedeiro.
 91
Bioorganic
5.2 Artigo: “Exploring the interaction between Chemistry
citrus 102 (2020) 104126
flavonoids and phytopathogenic fungi through
enzymatic activities”
Contents lists available at ScienceDirect

Bioorganic Chemistry
journal homepage: www.elsevier.com/locate/bioorg

T
Exploring the interaction between citrus flavonoids and phytopathogenic
fungi through enzymatic activities
Jonas Henrique Costa, Laura Soler Fernandes, Daniel Yuri Akiyama, Taícia Pacheco Fill
⁎

Institute of Chemistry, University of Campinas, CP 6154, 13083-970 Campinas, SP, Brazil

A R TICL E INFO A BSTR A CT

Keywords: Flavonoids are involved in citrus defense against phytopathogens. In this study, we applied in vitro biocatalysis
Biocatalysis assays using the flavanones glycosides hesperidin and naringin to explore the enzymatic activities involved in
Citrus such interaction. The main enzymatic activity observed was the hydrolysis catalyzed by fungi naringinases and
Flavonoid hesperidinases. Withing 7 days, the two citrus phytopathogenic fungi, Penicillium digitatum and Penicillium ita-
Fungus
licum, exhibited the highest hydrolyzing rate on the flavanones, reaching conversion values higher than 90%. In
Phytopathogen
Penicillium
addition, Geothrichum citri-aurantii exhibited no enzymatic activity and Penicillium expansum only hydrolyzed
hesperidin. In order to evaluate flavonoid biotransformation by the fungi in vivo, citrus fruits infected with P.
digitatum were analyzed through molecular networking and Imaging Mass Spectrometry (IMS). In vivo assays
revealed that citrus fruit in response to the infection is able to hydroxylate flavonoids, and novel flavonoid
structures were associated to the citrus’ defense. The data reported here present a new point of view in the
relation between citrus flavonoids and phytopathogenic fungi and can be useful to understand the infection
processes and host-pathogen interaction.

1. Introduction
Citrus have the largest global production compared to other fruit
genus and the forecast for citrus global production in 2019/2020 is
about 95 million metric tons (tons), including oranges, lemons, tan-
gerines and grapefruit [1]. Therefore, citriculture has an important
economic impact worldwide [2], especially for the largest producers
such as Brazil, China and United States [1].
Since citrus are rich in water and nutrients, these fruits are predis-
posed to be infected by phytopathogens during postharvest period [2]
and the fungal diseases are the main cause of citrus losses [3]. The main
fungi pathogens of citrus are Penicillium digitatum, Penicilium italicum
and Geotrichum citri-aurantii, which cause the green mold, blue mold
and sour rot diseases, respectively [2,4]. These diseases account for up
to 80% of the postharvest losses in citrus, leading to huge economic
losses [3].
Nowadays, the prevent of postharvest diseases caused by fungi in
citrus is through the use of chemical fungicides, such as imizalil,
leading to concerns about fungi resistance and environmental and
health risks [4,5]. In order to develop alternative and safer strategies to
protect citrus, it is necessary to elucidate the chemical interaction and
virulence strategies that occur during the host-pathogen interaction [6].
The interaction between plant and pathogen during the infection is
a battlefield in which the pathogen attacks and the plant activates de-
fense mechanisms [7]. For citrus fruit, it is well known the production
of phenylpropanoids as part of the defense mechanism [8,9,10]. The
citrus fruit peel is rich in flavanones, flavones and polymethoxylated
flavones, such as naringin, hesperidin, diosmin, tangeretin and nobi-
letin [8,11], that have been associated to defense against phytopatho-
gens, especially, P. digitatum [8,9,11]. Arcas et al. (2000) [10] reported
that fruit exposed to 2 h of UV light had the levels of naringin and
tangeretin increased by 7 and 55% respectively, and the growth of P.
digitatum in these fruits was reduced up to 45% when compared to non-
irradiated fruit. In another study, Kim et al. (2011) [9] showed that the
flavonoid level in fruit infected with P. digitatum decrease in the be-
ginning of the infection, then gradually increase in the intermediate
stage, before decreasing in the final stage. The initially decrease is due
to the known hydrolyzing action of P. digitatum that transform the
glycosylated flavanones to the corresponding aglycones [8,9,10,12],
while the increase in the intermediate infection stage, are associated to
the defense mechanisms of the plant [9].
Literature only describes the hydrolyzing ability of P. digitatum,
however other important citrus pathogens, such as P. italicum, have not
been studied concerning such enzymes and mechanisms. Furthermore,
other P. digitatum’s enzymatic activities on the citrus flavonoids were
never explored. Since the relation between the citrus phytopathogens

⁎
Corresponding author.
E-mail address: taicia@unicamp.br (T.P. Fill).

https://doi.org/10.1016/j.bioorg.2020.104126

Available online 22 July 2020

Received 14 May 2020; Received in revised form 29 June 2020; Accepted 20 July 2020

0045-2068/ © 2020 Elsevier Inc. All rights reserved.


J.H. Costa, et al.
and the flavonoids are important to understand the infection process,
the objective of this study was to compare the hydrolysis activity of the
main citrus pathogens (P. digitatum, P. citrinum, P. italicum and
Geotrichum citri-aurantii) and one non-citrus pathogen (P. expansum)
through in vitro biocatalysis using the main citrus flavonoids (hesper-
idin and naringin) as substrates. In addition, in vivo assays and mole-
cular networking was employed to investigate other biotransformations
in citrus flavonoids during host-pathogen interaction.
2. Material and methods
2.1. Commercial reagents
Commercially hesperidin, hesperitin, and naringin were purchased
from Sigma-Aldrich (St Luis, MO, USA).
2.2. Fungal strains and culture conditions
P. citrinum (PC) and Geotrichum citri-aurantii (GC) strains were ob-
tained from Sylvio Moreira Citrus Center (Cordeirópolis, SP, Brazil). P.
digitatum (PD) and P. italicum (PI) strains are deposited in the Spanish
Type Culture Collection (CECT) (accession code CECT20796 and
CECT20909, respectively). P. expansum (PE) are deposited in the
Tropical Culture Collection from André Tosello Collection (Campinas,
SP, Brazil) (accession code CCT 4680).
Fungi were cultured on potato dextrose agar (7 d at 25 °C). Spore
suspensions were prepared using sterile distilled water and were ad-
justed to a final concentration of 106 spores mL−1.
2.3. In vitro biotransformation of flavonoids
Erlenmeyer flasks (500 mL) containing 200 mL of potato dextrose
media and 40 µL of fungi spore suspension were incubated at 25 °C and
90 rpm during 72 h. The cells were vacuum filtered and washed with
acetate buffer (pH 4.0, 0.1 M).
The biocatalysis assays were performed using the protocol described
by Costa et al. (2017) [13] with few modifications. 1 g of wet cells,
20 mL of acetate buffer and 2 mg of flavonoid (diluted in 100 µL of
DMSO) were added to Erlenmeyer flasks (125 mL). Negative controls
were performed using acetate buffer and flavonoids to consider possible
spontaneous hydrolysis of the flavonoids. Also, fungi controls were
performed with acetate buffer and wet cells, to consider secondary
metabolites naturally produced by each of the fungi strains. All assays
and negative controls were performed in triplicate. The Erlenmeyer
flasks were incubated at 25 °C and 90 rpm during 7 d.
Assays and controls were monitored in the initial time and after 1, 3,
6, 24, 48, 72, 96 and 168 h (7 d) for PD and PI, and after 6 h for PC, PE
and GC. Aliquots of 1 mL were saturated with NaCl, extracted twice
with ethyl acetate (0.5 mL) and centrifuged at 18,000 g. Organic layers
were combined, dried, resuspended with 800 µL of HPLC methanol and
analyzed by HPLC-MS. Benzophenone was used as internal standard.
Enzymatic conversions (%C) were calculated using chromatographic
peak area of internal standard and reagent (Eq. (1)). Aisnc and Aisr are
the peak area of internal standard in the negative control and in the
reaction, respectively, and Arnc e Arr are the peak area of reagent in the
negative control and in the reaction, respectively.
%C = 100 100x (Arr xAisnc )/(Arnc xAisr ) (1)
2.4. In vivo assays
For in vivo assays, mature oranges (Citrus sinensis) purchased from a
local grocery store (Campinas, SP, Brazil) were surface sterilized and
wounded following the procedure described by Costa et al. (2019b)
[14]. Oranges were infected with 15 µL of P. italicum or P. digitatum
spore solution (106 spores mL−1). Infected and control oranges were
2
92
Bioorganic Chemistry 102 (2020) 104126
incubated in individual sterile 500 mL beakers (darkness at 25 °C)
during 3 or 5 d. Assays were performed in triplicate.
After the incubation period (3 or 5 d), the oranges peels (4 × 4 cm)
were cut on the infected area nearest the inoculation wound.
Extractions were performed following methodology described by
Smedsgaard (1997) [15] with few modifications. 0.6 g of orange peel
were extracted with 5 mL of methanol during 1 h in ultrasonic bath.
The obtained extracts were filtered, dried under N2 flux, resuspended in
1 mL of methanol HPLC and filtered through 0.22 µm into vials.
2.5. Mass spectrometry analysis
For in vitro assays, HPLC-MS analysis were performed using a Waters
ACQUITY UPLC system coupled to a Waters Micromass Quattro Micro
TM API, with electrospray ionization source and triple quadrupole mass
analyzer. Analyses were carried out in the positive mode with m/z
range of 100–1200, capillary voltage at 3 kV, cone voltage at 25 V, inlet
capillary temperature of 150 °C and nebulizing gas temperature of
200 °C. Stationary phase was Thermo Scientific column Accucore C18
(2.1 mm × 100 mm × 2.6 µm). Mobile phase was 0.1% formic acid (A)
and acetonitrile (B). Eluent profile (A/B): 95/5 up to 2/98 within
10 min, held for 5 min, up to 95/5 within 1.2 min and held for 3.8 min.
The total run time was 20 min for each run and the flow rate,
0.2 mL min−1. Injection volume was 5 µL.
For in vivo assays, high resolution mass spectrometry analyses
(HPLC-HRMS) were performed in a Waters Acquity UPLC H-class,
Waters Xevo G2-XS QToF mass spectrometer using electrospray ioni-
zation. Analyses were carried out in the positive mode with m/z range
of 100–1500, capillary voltage at 1.2 kV, source temperature at 100 °C;
cone gas (N2) flow of 50 L h−1 and dessolvation gas (N2) flow of 750 L
h−1. MS/MS were performed using collision energy ramp of 6–9 V (low
mass) and 60–80 V (high mass). Stationary phase was BEH C18 column
Accucore C18 (2.1 mm × 100 mm × 1.7 µm). Mobile phase was 0.1%
formic acid (A) and acetonitrile (B). Eluent profile (A/B): 90/10 up to
50/50 within 6 min, up to 2/98 within 3 min and up to 90/10 within
1 min. The total run time was 10 min for each run and the flow rate,
0.2 mL min−1. Injection volume was 2 µL. All the operation and spectra
analysis were conducted using Waters MassLynx (version 4.1).
2.6. Antifungal assays
For antifungal assays, 7 mL of PDA were supplemented with 20 µL
of hesperidin or hesperitin solution and poured in Petri dishes.
Flavonoid solutions were made with DMSO (0.3%) and final con-
centrations in PDA were 50, 200 and 400 mg L−1. 5 µL of spore solution
(106 spores mL−1) of PD, PI, PC, PE or GC were inoculated in on the
center of each agar plate. Negative controls were performed with 20 µL
of DMSO and the plates were incubated at 25 °C, in darkness, during 7
d.
2.7. Molecular network
A molecular networking for in vivo assays of P. digitatum was created
using the online workflow at GNPS (http://gnps.ucsd.edu). The data
was filtered by removing all MS/MS peaks within +/− 17 Da of the
precursor m/z. MS/MS spectra were window filtered by choosing only
the top 6 peaks in the +/− 50 Da window throughout the spectrum.
The data was then clustered with MS-Cluster with a parent mass tol-
erance of 0.04 Da and a MS/MS fragment ion tolerance of 0.04 Da to
create consensus spectra. Further, consensus spectra that contained less
than 2 spectra were discarded. A network was then created where edges
were filtered to have a cosine score above 0.5 and more than 4 matched
peaks. Further edges between two nodes were kept in the network only
if each of the nodes appeared in each other's respective top 10 most
similar nodes. The spectra in the network were then searched against
GNPS' spectral libraries. The library spectra were filtered in the same

J.H. Costa, et al.
manner as the input data. All matches kept between network spectra
and library spectra were required to have a score above 0.5 and at least
4 matched peaks [16].
2.8. Imaging mass spectrometry (IMS) analysis
For IMS, analyses were performed directly on orange peels infected
by P. digitatum at 4 and 6 days post inoculation (dpi), following pre-
viously protocol [14]. A negative control was also performed. Analyses
were carried out using a desorption electrospray ionization (DESI)
source Prosolia Model Omni Spray 2D®-3201) coupled to Thermo Sci-
entific QExactive® Hybrid Quadrupole-Orbitrap Mass Spectrometer.
Methanol flow rate was 10.0 mL min−1. Mass resolving power of
70.000 at m/z 200 was used to acquire IMS data. Bin width of Δm/
z ± 0.03 was used to generated images at Firefly data conversion
software (version 2.1.05). Images were visualized using Biomap soft-
ware (version 3.8.0.4) developed by Novartis Institutes for BioMedical
Research, and color scaling was fixed during the processing of each
image. MS spectra were processed using Xcalibur (version 3.0.63) de-
veloped by Thermo Fisher Scientific.
3. Results and discussion
3.1. In vitro biocatalysis: Hydrolysis of citrus flavonoids
Citrus flavonoids have been shown an important role during the
host-pathogen interaction [8,9,10,11,12]. As mentioned before, the
flavonoids act as a resistance barrier produced by the host against the
phytopathogens [8,9]; nevertheless, this defense mechanism is over-
come by the phytopathogens and the infections occur successfully in the
plants. Therefore, studies that evaluate the interaction between the
flavonoids and the phytopathogens are helpful to unravel the fungal
infection strategies and to understand the infection processes, leading
to the development of methods for controlling postharvest diseases.
In order to investigate the interaction of phytopathogens with the
citrus flavonoids during infection process, biocatalysis assays were
employed using hesperidin and naringin, as substrates, and whole cells
of the main citrus pathogens and one non-citrus pathogen as catalysts.
The use of whole cells in biocatalysis has advantages compared to
traditional chemical approaches as high selectivity, multi-step reactions
in a single strain, cofactor regeneration, high catalytic efficiency and it
is an environmentally friendly approach [17].
Reactions were monitored by HPLC-MS analyses (Table S1) (Figs.
S1–S6) and the main biotransformations detected were the hydrolysis of
hesperidin (1) in hesperitin-7-O-glucoside (2) followed by formation of
hesperitin (3), and the hydrolysis of naringin (4) in pruning (5) and in
the respective aglycone naringenin (6) (Fig. 1). For PD and PI, causal
agents of the most incident diseases in citrus, the reactions were
monitored from 0 to 168 h (7 d), since we expected faster and higher
hydrolysis rate for these fungi. PC, GC and PD were monitored after 6 h.
The enzymatic conversions (%) of 1 and 4 by fungi cells were cal-
culated using Eq. (1) and are represented in Figs. 2 and 3, respectively.
According to the graphs, obtained after monitoring the conversion rates
during 7 d, the phytopathogens have diverse biocatalytic pattern on
flavonoids, with some fungi reaching 100% of hydrolysis after 7 d
compared to 0% of conversion by others. Furthermore, comparing the
hydrolysis activity for the same strain on the different substrates 1 and
4, Figs. 2 and 3 also show different conversion profiles.
PD and PI showed similar profiles in the enzymatic hydrolysis of 1,
with high conversion at the beginning of the biocatalysis assays and
reaching 100% at 6 h. Apparently, PI demonstrated a higher conversion
rate of 1 when compared to PD at the first monitored time (77 versus
46%); however, until 6 h, most of the enzymatic product after hydro-
lysis of 1 by PD was compound 3 (Fig. S7 A) while for PI the major
compound was the product 2 (Fig. S7 B), therefore, most of the enzy-
matic activity by PI was the hydrolysis of the rhamnose portion of 1.
3
93
Bioorganic Chemistry 102 (2020) 104126
After 6 h, only the aglycone flavonoid 3 was detected for both fungi
(Fig. S8 A and B). PE and PC also exhibited similar profile in the hy-
drolysis of 1, reaching 40 and 80% of conversion, respectively, at 168 h
(Fig. S8 C and D). For PE and PC, compound 2 was not detected in the
assays. Lastly, GC did not show significantly enzymatic activity on 1,
with conversion rate close to zero during the experiment (Fig. S8 E).
PI exhibited the fastest and highest conversion rate of 4, reaching
100% of the aglycone product after 48 h (Fig. S9 A), however, unlike
the hydrolysis of compound 1 to 3 that started at the initial time of the
biocatalysis, the hydrolysis of flavanone 4 leading to compound 6
started at 3 h with 39% of conversion. PD showed another profile for
the hydrolysis of 4, with a conversion rate close to 0% at the first
monitored time which was gradually increasing to 91% at 168 h (Fig.
S9 B), unlike what was observed for 1, where PD reached 100% of
hydrolysis. Furthermore, the hydrolysis of 4 to 5 was only detected for
this fungus. PC reached 49% of hydrolysis of 4 at 168 h (Fig. S9 C) and
similar profile was observed for the hydrolysis of 1. GC and PE showed
a conversion rate for 4 close to 5% (Fig. S9 D and E), therefore, GC was
not able to effectively hydrolyze any of the flavanones, while PE only
exhibited high enzymatic activity on compound 1.
The enzymes responsible for the hydrolyzing action observed in vitro
are naringinase and hesperidinase, which are composed by two en-
zymes (α-L-rhamnosidase and β-glucosidase) [18,19,20,21,22,23]; α-L-
rhamnosidase catalyzes the hydrolysis of hesperidin to hesperitin-glu-
coside or of naringin to prunin, releasing rhamnose. Subsequently, β-
glucosidase hydrolyzes hesperitin-glucoside into hesperidin or prunin
into naringenin, releasing glucose [18,19,20,21,22,23]. In the litera-
ture, some plants, bacteria, yeast and filamentous fungi are reported as
naringinase and hesperidinases producers, mainly fungi from the genus
Aspergillus and Penicillium. Fungal naringinases and hesperidinases are
biotechnologically important and are applied in large amounts for in-
dustrial uses; these enzymes are employed, by example, to reduce the
bitterness of citrus fruit juice and to enhance wine aroma
[18,19,20,21,22,23]. Our data showed that P. italicum is a promising
new source of naringinase and hesperidinase, since it exhibited faster
and higher hydrolytic activity when compared with other fungi.
The hydrolyzing action of P. digitatum on citrus flavonoids were
already known in the literature [10,12]. Furthermore, López-Peréz
et al. (2015) [24] investigated P. digitatum genes expressed during in-
fection in citrus by RNAseq. The authors reported that genes of hy-
drolase activities, such as hydrolysis of O-glycosyl compounds, were up-
regulated during the infection process and the functions encoded by
these up-regulated genes are related to a pathogenic life style [24].
However, although the hydrolyzing activity has an important role for
virulence, it is the first time that this enzymatic activity is measured for
P. digitatum and compared with other main citrus pathogens. The in
vitro assays showed different profiles for the hydrolysis of naringin and
hesperidin by P. digitatum, which deserves future research to reveal how
this affect the infection process and host-pathogen interaction.
The hydrolysis activity detected for the most of the phytopathogens
contradicts data reported by previous studies, which observed that the
aglycons compounds have more antifungal action than the glycosylated
flavanones [12]. Ortuño and Del Río (2009) [12] reported that 3 and 6
are more active fungistatic agents against P. digitatum than 1 and 4. The
concentration (mM) of 1, 3, 4 and 6, in PDA medium, to inhibit 50% of
P. digitatum radial growth (IC50) was 7.86 ± 0.3, 0.12 ± 0.01,
17.9 ± 2.3 and 0.32 ± 0.02, respectively. To confirm this data, we
performed antifungal assays using PDA medium supplemented with 1
or 3 however, no significant difference was observed in fungal radial
growth between 1 and 3 at the concentrations tested (50, 100 or
400 mg L−1). (Fig. S10) which may indicate similar growth pattern in
the presence of both 1 and 3. Therefore, the difference in the antifungal
potential between the flavonoids need to be revised and deserves fur-
ther investigation.
In contrast to P. digitatum, G. citri-aurantii did not hydrolyze any of
the flavonoids, nevertheless the sour rot has been reported as a serious

J.H. Costa, et al.
Fig. 1. Hydrolysis of hesperidin (1) and naringin (4) by citrus pathogens.
Fig. 2. Enzymatic conversion (%) of 1 at different time intervals by PD, PI, GC,
PC and PE. Values are expressed as the mean (n = 3) ± standard deviation
(vertical bars).
Fig. 3. Enzymatic conversion (%) of 4 at different time intervals by PD, PI, GC,
PC and PE. Values are expressed as the mean (n = 3) ± standard deviation
(vertical bars).
4
94
Bioorganic Chemistry 102 (2020) 104126
citrus disease, difficult to control because of the lack of effective fun-
gicides [4,25]. RNA-seq analysis revealed that in citrus fruit infected by
P. digitatum, P. italicum or G. citri-aurantii the enzymes involved in fla-
vonoids biosynthesis are up-regulated [26]. Other study reported that
G. citri-aurantii mycelial growth was inhibited up to 54% by essential oil
of oranges containing polymethoxylated flavonoids [27]. Thus, the in-
teraction between citrus flavonoids and G. citri-aurantii seems to be
important during the infection process as for P. digitatum. Pre-harvest
fungi have also been studied concerning the biotransformation of fla-
vonoids. Díaz et al. (2015) [28] reported that Alternaria alternata, a pre-
harvest citrus pathogen, degrades citrus flavonoids. Flavonoids were
added in liquid culture of A. alternata and, after 4 d, concentration of
hesperidin, naringenin and diosmin had fallen 50, 75 and 99%, re-
spectively [28]. However, the degradative metabolism by A. alternata is
through extracellular laccases and no β-glucosidase activity was ob-
served [28]. Thus, even though G. citri-aurantii did not exhibit rham-
nosidase or glucosidase activities, we speculate that G. citri-aurantii
interacts with citrus flavonoids through other enzymatic activities.
Therefore, since the mechanisms of virulence were not fully revealed
for this fungus and there is a lack of information in the literature, the
data reported here open a field of investigation for the sour rot caused
by G. citri-aurantii.
In addition, regarding to the role of hydrolytic enzymes during the
infection, Lafuente et al. (2019) [29] reported that the phytohormone
abscisic acid (ABA) is involved in the protection of the fruit against P.
digitatum. In the early stage of the infection, ABA levels decrease after P.
digitatum infection and a rise level is observed in the later infection
time. One of the hypotheses raised by the authors is that the ABA in-
creasing is consequence of releasing free ABA from the glucose-con-
jugated ABA (ABA-GE) and the increase in ABA level in later infection
might contribute to fungal colonisation. The authors suggest that fungal
β-glucosidases could be involved in this process, but more studies are
necessary to confirm this hypothesis and to describe the role of P. di-
gitatum β-glucosidases [29].
The fact that P. expansum, a non-citrus pathogen, is able to hydro-
lyze hesperidin and the non-enzymatic activity showed by G. citri-aur-
antii are in agreement with the literature: the hydrolysis of plant de-
fense molecules is not the unique virulence factor existing during the
infection process [6]. Ciegler et al. (1971) [30], studied the transfor-
mation of flavonoids naringin and naringenin by various fungi strains.
 95
J.H. Costa, et al. Bioorganic Chemistry 102 (2020) 104126

Fig. 4. Clusters A, B and C of MS/MS network analysis of PD infected and control oranges extracts. Nodes in bold blue represent GNPS hits. Marked carbons 6 and 8 in
red text in 12 and 13 represent possible positions of a common hydroxyl group, correct position was not able to be determined through MS/MS analysis. (For
interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

The author observed that P. expansum was unable to act on naringin,
although it hydrolyzed hesperidin. Furthermore, the authors only ob-
served the hydrolysis activity and no transformation of aglycone fla-
vonoids were detected [30]. These results are in agreement with our
data, since we did not detect any other enzymatic activity in vitro.
Ciegler et al. (1971) [30] attributes this inability to the fact that these
flavonoids are relatively insoluble and impermeable into cells.
Naringinases e hesperidinases have not been related to the patho-
genicity in the literature. Penicillium species have been reported as
naringinase producers [31], however, a number of publications focus
on the immobilization of naringinases and application in industries
[32]. Yet, Penicillium species also have been reported to have a high
capacity to produce extracellular β-glucosidades [32,33]. Some β-glu-
cosidades have been related as part of fungal virulence strategy, since
fungal pathogens secrete β-glucosidades to break the cellulose
molecules of plant cell wall to facilitate their penetration [34]. β-glu-
cosidades are part of cell wall degrading enzymes (CWDEs) and during
P. digitatum infection in citrus, CWDEs genes were up-regulated and
lacking mutants in genes enconding CDWEs were less virulent, sup-
porting the role in pathogenicity of specific CWDEs [24]. Nevertheless,
the action of these enzymes on citrus defense metabolites, regarding to
virulence is not well established.
3.2. In vivo biotransformation assays
3.2.1. Hydrolysis of citrus flavonoids
For in vivo assays, PD and PI were selected since they presented
higher conversion rate in vitro and are the main citrus pathogens re-
ported in the literature [2,4]. Similarly as the in vitro experiments, the
main biotransformations observed were the hydrolysis of 1 into 2 and

5

J.H. Costa, et al.
Fig. 5. Fragmentation for flavanones 3, 11, 12 and 13. Possible hydroxylation positions are red marked carbons 6 and 8. (For interpretation of the references to color
in this figure legend, the reader is referred to the web version of this article.)
3, and the hydrolysis of 4 into 5 and 6. The infected fruit were mon-
itored by HPLC-MS analysis at 3 and 5 days post inoculation (dpi). The
enzymatic conversation (%) of 1 and 4 was calculated using Equation
(1).
PI and PD exhibited close conversion rate for compound 1 and 4 at
the early stage of the infection. At 3 dpi, the conversion rate for 1 was
0.97% by PD and 0,90% by PI, while for 4, PD and PI hydrolyzed 2.08
and 3.82%, respectively. However, at the later infection, PI showed the
highest conversion rates for the flavanones in vivo, when compared with
PD. At 5 dpi, PI hydrolyzed 6.43 and 17.08% of compounds 1 and 4,
respectively, while PD exhibited conversion rate of 1.56% for com-
pound 1 and 2.56% for compound 4.
As mentioned before, flavonoids in citrus species act in the defense
mechanism against fungi cells, and their composition can change
during pathogen attack [8,9]. In the in vitro study (section 3.1), it was
observed that PD and PI totally hydrolyzed the glycoside flavonoids in
its aglycone forms (Figs. 2 and 3). Therefore, it was expected that in the
in vivo studies, this result would remain. Nevertheless, in vivo assays
showed that, although the enzymatic conversion had increase over
time, the percentage of this conversion was below expectation for both
PD and PI. This can be related to the fact that, while the fungus hy-
drolyzes the flavonoids, the fruit produces more compounds to defend
itself from attack [8,9]. Kim et al. (2011) [9] measured the level of 1
and 4 in oranges infected by P. digitatum and reported that the levels of
these flavonoids decrease at the early stage of infection, gradually in-
crease in the middle stage and subsequent decrease at later infection.
The biosynthesis of the defense mechanisms by the fruit competes with
de fungus metabolism to consume them [9], which can explain the data
observed here for the in vivo assays and that it is hard to measure the
6
96
Bioorganic Chemistry 102 (2020) 104126
hydrolysis rate of flavonoids in vivo. Thus, while the fungus decreases
the levels of 1 and 4, converting into the aglycones species 3 and 6
respectively, the citrus fruit produce more 1 and 4, decreasing the en-
zymatic conversion calculated.
3.2.2. Biotransformation of flavonoids in vivo by P. digitatum: A molecular
network approach
To investigate in vivo hydrolysis of 1 and 4 and other bio-
transformation products present in this interaction, molecular net-
working was applied as a dereplication tool for fruit infected by P. di-
gitatum to further characterize infection’s metabolic profile.
Analyses of metabolite’s fragmentation patterns, acquired through
tandem MS techniques, and comparison with databases have become a
standard procedure in natural product discovery and other metabolic
studies [35]. In this study, we utilized GNPS platform (Global Natural
Products Social Molecular Networking), one of the largest natural
products public database with regards to MS/MS spectra [16].
In molecular networking analyses, each node represents one meta-
bolite, labeled according to its m/z ratio. Based on similar fragmenta-
tion patterns, GNPS platform is able to group nodes that represent
molecules with similar structures in clusters [16]. GNPS dereplication
with extracts from both PD infected and control orange fruit resulted in
a network containing 149 clusters. For the purpose of this study, clus-
ters A, B and C (Fig. 4) were chosen for deeper analysis since they al-
lowed visualization of the main citrus flavonoids. HRESI-MS data of all
compounds reported in this study are compiled in Table S2.
Based on structure similarity, molecular network grouped in cluster
A flavonoids bonded to disaccharides: hesperidin (1), naringin (4) and
didymin (7), were identified through GNPS MS/MS library (Figs.

J.H. Costa, et al.
S11–S13) and are flavonoids normally found in citrus fruit [36]. Cluster
B identified (Figs. S14–S16) and grouped products with a single sac-
charide: hesperitin-7-O-glucoside (2), pruning (5) and luteolin-7-O-
glucoside (8), also compounds that can be commonly found in citrus
[36].
Interestingly, cluster C presented and identified the products of total
hydrolysis of 1 and 4: hesperitin (3) and naringenin (6) (Figs. S17 and
S18). Also, other aglycone flavonoids such as isosakuranetin (9) (Fig.
S19), eriodictyol (10) (Fig. S20) and diosmetin (11) were grouped in
this cluster. The presence of this compounds in the molecular network
was expected, since they are present in citrus fruit [36]. However,
compounds 12 and 13 were only found in the infected fruit and were
not identified as citrus flavonoids. Both possible isomers of 12 have
already been described in the literature, but never in citrus [37,38].
Based on fragmentation profile of 12 and 13 and their comparison
to 3 and 11, it was possible to confirm the presence of a hydroxyl group
in 12 and 13 in the A ring of the flavanone structure (Fig. 5). Flava-
nones present a common fragmentation profile containing retro-
cyclization cleavages of the C-ring [39]. In the fragmentation spectra of
3 (Fig. S21) and 11 (Fig. S22), fragment m/z 153 is present, although it
is absent in the MS/MS spectra of 12 (Fig. S23) and 13 (Fig. S24). In-
stead, a new fragment corresponding to m/z 169 is identified. The m/z
difference of these fragments suggests the presence of a hydroxyl group
in ring A of 12 and 13, however the correct position of the hydroxyl
group cannot be determined by mass spectrometry.
As mentioned before, flavonoids 12 and 13 were only present in PD
infected oranges, suggesting that these molecules are biotransformation
products of other citrus flavonoids. Based on this data, two hypotheses
are possible: biotransformation is carried out by either (i) PD as an
infection mechanism or (ii) citrus as a defense mechanism.
Flavonoids 12 and 13 were not found in the in vitro assays, in-
dicating that their production may be part of citrus’ defense mechan-
isms. Furthermore, based on structure similarity, 12 and 13 are the
hydroxylation products of 11 and 3, respectively. Studies analyzing P.
digitatum’s infection in citrus through transcriptomic approaches reveal
upregulation of citrus’ genes related to O-methyltransferases, cyto-
chrome P450 monooxygenases and hydroxylases, as well as several
genes that participate in the flavonoid biosynthesis pathway [40]. Thus,
we hypothesize that 12 and 13 are citrus biotransformation products as
a defensive mechanism to P. digitatum’s infection.
To confirm our hypothesis, IMS analyses were performed in oranges
infected with P. digitatum at 4 and 6 dpi (Fig. S25). IMS is a powerful
tool, since an image is generated for each ion detected on the analyzed
surface, correlating molecules and their spatial distribution [41]. In
previous studies, IMS technique was successfully applied in oranges
with green mold disease to monitor the infection process and the pro-
duction of secondary metabolites by P. digitatum [14]. Fig. 6 shows IMS
signals obtained for ion [M + H]+ 319.0813 (Fig. S26), which corre-
spond to compound 13 (error = 0.2 ppm). IMS images revealed that
compound 13 is accumulated in the infected oranges when compared to
the control, yet, 13 is intensively concentrated on the peel about to be
infected by the fungus, confirming that compounds 12 and 13 are part
of the plant defense mechanism. It is the first time that these com-
pounds are reported as a citrus defense mechanism against phyto-
pathogens and further investigation is needed to describe the role of
these compounds in citrus defense.
7
97
Bioorganic Chemistry 102 (2020) 104126
Fig. 6. Imaging mass analyses obtained by (+)
DESI-MS of oranges infected by P. digitatum at 4
and 6 dpi. Each MS image shows the spatial
distribution of m/z 319.08 over orange peel. All
images are plotted on the same color scale from
0 (black) to 5x104 (red). (For interpretation of
the references to color in this figure legend, the
reader is referred to the web version of this ar-
ticle.)
4. Conclusions
Among the molecular factors that occur during the infection of ci-
trus fruit by phytopathogens, the involvement of the flavonoids is one
of the most known and yet not fully explored. As far as we know, this is
the first report to study the enzymatic activity of phytopathogenic fungi
on citrus flavanones. The in vitro biocatalysis assays employing nar-
ingin, hesperidin and whole fungi cells demonstrated to be an appro-
priate technique to measure and detect enzymatic activities. According
to our data, the hydrolysis activity is different among fungi, with P.
digitatum and P. italicum showing high conversion rates and G. citri-
aurantii exhibiting no enzymatic activity. Also, through in vivo assays,
molecular networking and IMS, our experiments demonstrated, for the
first time, novel compounds involved in defense of citrus against phy-
topathogens. Lastly, this study opens new opportunities of research
since our data add more information about the fungal response against
the plant defense molecules, and the enzymes described here are
commercial and biotechnologically important.
Funding
This work was supported by the Coordenação de Aperfeiçoamento
de Pessoal de Nível Superior - Brasil (CAPES) [Finance Code 001],
Fundação de Amparo a Pesquisa no Estado de São Paulo [grant numbers
2018/13027–8, 2019/06359–7, 17/24462–4] and L’Oréal Brazil, to-
gether with ABC and UNESCO in Brazil.
CRediT authorship contribution statement
Jonas Henrique Costa: Conceptualization, Investigation, Writing -
original draft. Laura Soler Fernandes: Investigation, Writing - original
draft. Daniel Yuri Akiyama: Investigation, Writing - original draft.
Taícia Pacheco Fill: Conceptualization, Methodology, Resources,
Supervision, Writing - review & editing.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
Acknowledgements
We would like to thank Dr. Katia Kupper for P. citrinum and
Geotrichum citri-aurantii strains, and Dr. Ljubica Tasic for the commer-
cially flavonoids. The authors thank Coordenação de Aperfeiçoamento
de Pessoal de Nível Superior - Brasil (CAPES) [Finance Code 001],
Fundação de Amparo a Pesquisa no Estado de São Paulo [grant numbers
2018/13027–8, 2019/06359–7, 17/24462–4] and L’Oréal Brazil, to-
gether with ABC and UNESCO in Brazil for the funding.
Appendix A. Supplementary material
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bioorg.2020.104126.

J.H. Costa, et al.
References
[1] Citrus: World Markets and Trade. United States Department of Agriculture. Foreign
Agricultural Service, January, p. 1–13, 2020. Available online: https://downloads.
usda.library.cornell.edu/usda-esmis/files/w66343603/00000g55g/kp78h0193/ci-
trus.pdf (accessed 14 May 2020).
[2] I. Talibi, H. Boubaker, E.H. Boudyach, A.B. Aoumar, Alternative methods for the
control of postharvest citrus diseases, J. Appl. Microbiol. 117 (1) (2014) 1–17,
https://doi.org/10.1111/jam.12495.
[3] M. El-Otmani, A. Ait-Oubahou, L. Zacarías, Citrus spp.: Orange, mandarin, tan-
gerine, clementine, grapefruit, pomelo, lemon and lime, Postharvest Biol. Technol.
Tropical Subtropical Fruits 437–514 (2011) 515e–516e, https://doi.org/10.1533/
9780857092762.437.
[4] J.M. Bazioli, J.R. Belinato, J.H. Costa, D.Y. Akiyama, J.G.M. Pontes, K.C. Kupper,
F. Augusto, J.E. Carvalho, T.P. Fill, Biological control of citrus phytopathogens,
Toxins 11 (2019) 460, https://doi.org/10.3390/toxins11080460.
[5] M. Ramón-Carbonell, P. Sánchez-Torres, Significance of 195 bp-enhancer of
PdCYP51B in the acquisition of Penicillium digitatum DMI resistance and increase of
fungal virulence, Pestic. Biochem. Physiol. (2020), https://doi.org/10.1016/j.
pestbp.2020.01.003.
[6] J.H. Costa, J.M. Bazioli, J.G. Ponte, T.P. Fill, Penicillium digitatum infection me-
chanisms in citrus: What do we know so far? Fungal Biol. 123 (8) (2019) 584–593,
https://doi.org/10.1016/j.funbio.2019.05.004.
[7] K. Shirasu, R. Maor, The arms race continues: battle strategies between plants and
fungal pathogens, Curr. Opin. Microbiol. 8 (4) (2005) 399–404, https://doi.org/10.
1016/j.mib.2005.06.008.
[8] A. Ortuño, L. Díaz, N. Alvarez, I. Porras, A. García-Lidón, J.A. Del Río, Comparative
study of flavonoid and scoparone accumulation in different Citrus species and their
susceptibility to Penicillium digitatum, Food Chem. 125 (1) (2011) 232–239, https://
doi.org/10.1016/j.foodchem.2010.09.012.
[9] H.G. Kim, G. Kim, J.H. Lee, S. Park, W.Y. Jeong, Y.H. Kim, S.T. Kim, Y.A. Cho,
W.S. Lee, S.J. Lee, J.S. Jin, S.C. Shin, Determination of the change of flavonoid
components as the defence materials of Citrus unshiu Marc. fruit peel against
Penicillium digitatum by liquid chromatography coupled with tandem mass spec-
trometry, Food Chem. 128 (1) (2011) 49–54, https://doi.org/10.1016/j.foodchem.
2011.02.075.
[10] M.C. Arcas, J.M. Botía, A.M. Ortuño, J.A. Del Rio, UV irradiation alters the levels of
flavonoids involved in the defence mechanism of Citrus aurantium fruits against
Penicillium digitatum, Eur. J. Plant Pathol. 106 (2000) 617–622.
[11] A. Ortuño, A. Baídez, P. Gómez, M.C. Arcas, I. Porras, A. García-Lídon, J.A. Del Río,
Citrus paradisi and Citrus sinensis flavonoids: Their, influence in the defence me-
chanism against Penicillium digitatum, Food Chem. 98 (2) (2006) 351–358, https://
doi.org/10.1016/j.foodchem.2005.06.017.
[12] A. Ortuño, J.A. Del Río, Role of Citrus phenolic compounds in the resistance me-
chanism against pathogenic fungi, Tree Forestry Sci. Biotechnol. 3 (Special Issue 2)
(2009) 49–53.
[13] J.H. Costa, B.Z. Costa, D.A. Angelis, A.J. Marsaioli, Monoamine oxidase and
transaminase screening: biotransformation of 2-methyl-6-alkylpiperidines by
Neopestalotiopsis sp. CBMAI 2030, Appl. Microbiol. Biotechnol. 101 (2017)
1061–1070, https://doi.org/10.1007/s00253-017-8389-z.
[14] J.H. Costa, J.M. Bazioli, E.V. Araújo, P.H. Vendramini, M.C.F. Porto, M.N. Eberlin,
J.A. Souza-Neto, T.P. Fill, Monitoring indole alkaloid production by Penicillium di-
gitatum during infection process in citrus by Mass Spectrometry Imaging and mo-
lecular networking, Fungal Biol. 123 (8) (2019) 594–600, https://doi.org/10.1016/
j.funbio.2019.03.002.
[15] J. Smedsgaard, Micro-scale extraction procedure for standardized screening of
fungal metabolite production in cultures, J. Chromatogr. 760 (1997) 264.
[16] M. Wang, et al., Sharing and community curation of mass spectrometry data with
global natural products social molecular networking, Nat. Biotechnol. 34 (2016)
828–837 PMID: 27504778.
[17] Y. Tao, B. Lin, Whole-cell biocatalysts by design, Microb. Cell Fact. 16 (2017) 106,
https://doi.org/10.1186/s12934-017-0724-7.
[18] Y. Zhu, H. Jia, M. Xi, J. Li, L. Yang, X. Li, Characterization of naringinase from
Aspergillus oryzae 11250 and its application in the debitterization of orange juice,
Process Biochem. 62 (2017) 114–121, https://doi.org/10.1016/j.procbio.2017.07.
012.
[19] S. Yadav, R.S.S. Yadav, K.D.S. Yadav, An α-L-rhamnosidase from Aspergillus awamori
MTCC-2879 and its role in debittering of orange juice, Int. J. Food Sci. Technol. 48
(2013) 927–933, https://doi.org/10.1111/ijfs.12043.
[20] J. Bodakowska-Boczniewicz, Z. Garncarek, Immobilization of naringinase from
Penicillium decumbens on chitosan microspheres for debittering grapefruit juice,
Molecules 24 (23) (2019) 4234, https://doi.org/10.3390/molecules24234234.
8
98
Bioorganic Chemistry 102 (2020) 104126
[21] R.S. Singh, T. Singh, A. Pandey, Microbial enzymes – an overview, Biomass,
Biofuels Biochem.: Adv. Enzyme Technol. 1–40 (2019), https://doi.org/10.1016/
B978-0-444-64114-4.00001-7.
[22] J. Wang, Y. Ma, X. Wu, L. Yu, R. Xia, G. Sun, F. Wu, Selective hydrolysis by com-
mercially available hesperidinase for isoquercitrin production, J. Mol. Catal. B
Enzym. 81 (2012) 37–42, https://doi.org/10.1016/j.molcatb.2012.05.005.
[23] A.F.M. Furtado, M.A.P. Nunes, M.H.L. Ribeiro, Hesperidinase encapsulation to-
wards hesperitin production targeting improved bioavailability, J. Mol. Recognit.
25 (2012) 595–603, https://doi.org/10.1002/jmr.2224.
[24] M. López-Pérez, A. Ballester, L. González-Candelas, Identification and functional
analysis of Penicillium digitatum genes putatively involved in virulence towards ci-
trus fruit, Mol. Plant Pathol. 16 (3) (2015) 262–275, https://doi.org/10.1111/mpp.
12179.
[25] Y. Wu, X. Duan, G. Jing, O.Q. Yang, N. Tao, Cinnamaldehyde inhibits the mycelial
growth of Geotrichum citri-aurantii and induces defense responses against sour rot in
citrus fruit, Postharvest Biol. Technol. 129 (2017) 23–38, https://doi.org/10.1016/
j.postharvbio.2017.03.004.
[26] B. Deng, W. Wang, L. Deng, S. Yao, J. Ming, K. Zeng, Comparative RNA-seq analysis
of citrus fruit in response to infection with three major postharvest fungi,
Postharvest Biol. Biotechnol. 146 (2018) 134–146, https://doi.org/10.1016/j.
postharvbio.2018.08.012.
[27] J.A. Del Río, M.C. Arcas, O. Benavente-García, A. Ortuño, Citrus polymethoxylated
flavones can confer resistance against Phytophthora citrophthora, Penicillium digi-
tatum, and Geotrichum species, J. Agric. Food. Chem. 46 (10) (1998) 4423–4428,
https://doi.org/10.1021/jf980229m.
[28] L. Díaz, J.A. Del Río, M. Pérez-Gilabert, A. Ortuño, Involvement of an extracellular
fungus laccase in the flavonoid metabolism in Citrus fruits inoculated with
Alternaria alternata, Plant Physiol. Biochem. 89 (2015) 11–17, https://doi.org/10.
1016/j.plaphy.2015.02.006.
[29] M.T. Lafuente, A. Ballester, L. González-Candelas, Involvement of abscisic acid in
the resistance of citrus fruit to Penicillium digitatum infection, Postharvest Biol.
Technol. 154 (2019) 31–40, https://doi.org/10.1016/j.postharvbio.2019.04.004.
[30] A. Ciegler, L.A. Lindenfelser, G.E.N. Nelson, Microbial transformation of flavonoids,
Appl. Microbiol. 22 (1971) 974–979.
[31] M.H. Ribeiro, Naringinases: occurrence, characteristics, and applications, Appl.
Microbiol. Biotechnol. 90 (6) (2011) 1883–1895, https://doi.org/10.1007/s00253-
011-3176-8.
[32] G. Molina, E.A. Lima, G.P. Borin, M.C.S. Barcelos, G.M. Pastore, Beta-glucosidase
from Penicillium, New and Future Developments in Microbial Biotechnology and
Bioengineering: Penicillium System Properties and Applications, Elsevier, 2018, pp.
137–151, , https://doi.org/10.1016/B978-0-444-63501-3.00007-7.
[33] D. Mamma, E. Kalogeris, D.G. Hatzinikolaou, A. Lekanidou, D. Kekos, B.J. Macris,
P. Christakopoulos, Biochemical characterization of the multi-enzyme system pro-
duced by Penicillium decumbers grown on rutin, Food Biotechnol. 18 (1) (2004)
1–18, https://doi.org/10.1081/FBT-120030382.
[34] P.K. Sudheeran, R. Ovadia, O. Galsarker, I. Maoz, N. Sela, D. Maurer,
O. Feygenberg, M.O. Shamir, N. Alkan, Glycosylated flavonoids: fruit’s concealed
antifungal agents, New Phytol. 225 (2019) 1788–1798, https://doi.org/10.1111/
nph.16251.
[35] B.C. Covington, J.A. McLean, B.O. Bachmann, Comparative mass spectrometry-
based metabolomics strategies for the investigation of microbial secondary meta-
bolites, Nat. Prod. Rep. 34 (1) (2017) 6–24, https://doi.org/10.1039/c6np00048g.
[36] G. Gattuso, D. Barreca, C. Gargiulli, U. Leuzzi, C. Caristi, Flavonoid composition of
citrus juices, Molecules 12 (8) (2007) 1641–1673, https://doi.org/10.3390/
12081641.
[37] D.C. Albach, R.J. Grayer, S.R. Jensen, F. Özgökce, N.C. Veitch, Acylated flavone
glycosides from Veronica, Phytochemistry 64 (7) (2003) 1295–1301, https://doi.
org/10.1016/j.phytochem.2003.08.012.
[38] A. Lenherr, M.F. Lahloub, O. Sticher, Three flavonoid glycosides containing
acetylated allose from Stachys Recta, Phytochemistry 23 (10) (1984) 2343–2345,
https://doi.org/10.1016/S0031-9422(00)80548-8.
[39] D. Tsimogiannis, M. Samiotaki, G. Panayotou, V. Oreopoulou, Characterization of
flavonoid subgroups and hydroxy substitution by HPLC-MS/MS, Molecules 12 (3)
(2007) 593–606, https://doi.org/10.3390/12030593.
[40] L. González-Candelas, S. Alamar, P. Sanchez-Torres, L. Zacarias, J.F. Marcos, A
transcriptomic approach highlights induction of secondary metabolism in citrus
fruit in response to Penicillium digitatum infection, BMC Plant Biol. 10 (1) (2010)
194, https://doi.org/10.1186/1471-2229-10-194.
[41] S. Bertrand, N. Bohni, S. Schnee, O. Schumpp, K. Gindro, J. Wolfender, Metabolite
induction via microorganism co-culture: A potential way to enhance chemical di-
versity for drug discovery, Biotechnol. Adv. 32 (6) (2014) 1180–1204, https://doi.
org/10.1016/j.biotechadv.2014.03.001.
 99

5.3 Informações suplementares do capítulo III

Exploring the interaction between citrus flavonoids and phytopathogenic fungi

through enzymatic activities

Jonas Henrique Costaa, Laura Soler Fernandesa, Daniel Yuri Akiyamaa, Taícia
Pacheco Filla*

a Institute of Chemistry, University of Campinas, CP 6154, 13083-970, Campinas –

SP, Brazil

*Corresponding author: phone +55-19-35213092, e-mail taicia@unicamp.br

Supplementary data

Table S1. HPLC-MS data obtained for compounds 1 – 6.

Compound Name [M+H]+ m/z Retention time (min)
1 hesperidin 611 3.7
2 hesperitin-7-O-glucoside 465 3.9
3 hesperitin 303 5.3
4 naringin 581 3.7
5 pruning 435 3.8
6 naringenin 273 5.1
 100

Figure S1. Mass spectrum of ion [M+H]+ m/z 611 obtained for hesperidin (1) at 3.7 min.

Figure S2. Mass spectrum of ion [M+H]+ m/z 465 obtained for hesperitin-7-O-glucoside (2) at 3.9 min.

Figure S3. Mass spectrum of ion [M+H]+ m/z 303 obtained for hesperitin (3) at 5.3 min.

Figure S4. Mass spectrum of ion [M+H]+ m/z 581 obtained for naringin (4) at 3.7 min.
 101

Figure S5. Mass spectrum of ion [M+H]+ m/z 435 obtained for pruning (5) at 3.8 min.

Figure S6. Mass spectrum of ion [M+H]+ m/z 273 obtained for naringenin (6) at 5.1 min.

Figure S7. Extracted ion chromatograms of m/z 303 for in vitro biotransformation of hesperidin (1) by
A) PD and B) PI at the initial time of the biocatalysis assays.
 102

Figure S8. Extracted ion chromatograms of m/z 303 for in vitro biotransformation of hesperidin (1) by
A) PI, B) PD, C) PC, D) PE and E) GC at 168 h (7 days). Chromatogram F): control of 1.
 103

Figure S9. Extracted ion chromatograms of m/z 273 for in vitro biotransformation of naringin (4) by A)
PI, B) PD, C) PC, D) PE and E) GC at 168 h (7 days). Chromatogram F): control of 4.
 104

Figure S10. Phytopathogenic fungi growth in PDA medium supplemented with hesperidin and
hesperitin (400 mg L-1) after 96h.

Table S2. HRESI-MS data obtained for compounds 1 – 13.

Compound Ion formula Calculated Experimental Error (ppm)

([M+H]+) m/z m/z
1 C28H35O15 611.1976 611.1977 0.2
2 C22H25O11 465.1397 465.1416 4.1
3 C16H15O6 303.0869 303.0884 4.9
4 C27H33O14 581.1870 581.1882 2.1
5 C21H23O10 435.1291 435.1298 1.6
6 C15H13O5 273.0763 273.0762 -0.4
7 C28H35O14 595.2027 595.2029 0.3
8 C21H20O11 448.1006 448.1009 0.7
9 C16H15O5 287.0918 287.0927 3.1
10 C15H13O6 289.0712 289.0723 3.8
11 C16H13O6 301.0712 301.0714 0.7
12 C16H13O7 317.0661 317.0663 0.6
13 C16H15O7 319.0818 319.0811 -2.2
 110

Figure S21. MS/MS (+) spectra of hesperitin (3).

Figure S22. MS/MS (+) spectra of diosmetin (11).

 111

Figure S23. MS/MS (+) spectra of compound 12.

Figure S24. MS/MS (+) spectra of compound 13.

 112

Figure S25. Oranges fruits (Citrus sinensis) analyzed by Imaging Mass Spectrometry (IMS).

150 #75 RT: 0.34 AV: 1 NL: 2.64E4

T: FTMS + p NSI Full ms [100.0000-1500.0000]
319.0813
C 16 H 15 O 7
0.2181 ppm
100

90

80

70

60

50

40

30

20

10

0
319.06 319.07 319.08 319.09 319.10
m/z

Figure S26. Mass spectrum of ion [M+H]+ m/z 319.0813 obtained for compound (13) through DESI-
IMS.
 113

6 CAPÍTULO IV
Potencial antifúngico de metabólitos secundários envolvidos na interação
entre patógenos cítricos

O conteúdo deste capítulo é composto pelo artigo intitulado “Antifungal potential

of secondary metabolites involved in the interaction between citrus
pathogens”, publicado no periódico Scientific Reports. A reprodução pelos
autores é autorizada sob os termos da Creative Commons Attribution 4.0
International license (anexo 10.2).

Referência: Costa, J.H., Wassano, C.I., Angolini, C.F.F., et al. 2019. Antifungal
potential of secondary metabolites involved in the interaction between citrus
pathogens. Scientific Reports, 9:18647. doi: 10.1038/s41598-019-55204-9

Versão final publicada disponível em:

https://www.nature.com/articles/s41598-019-55204-9

6.1 Resumo
Considerando-se os métodos atuais de controle de fitopatógenos através
de fungicidas sintéticos, neste projeto a técnica de co-cultivo foi utilizada em conjunto
com IMS na busca de fungicidas naturais que possam ser utilizados em substituição
aos fungicidas químicos que trazem preocupações em relação à saúde humana e ao
meio ambiente. MSI foi empregado em co-cultura de P. citrinum e P. digitaum, dois
fitopatógenos que competem pelo mesmo hospdeiro. Através da técnica de MSI foi
possível visualizar uma guerra química entre os dois fungos: dois tetrapeptídeos,
deoxicitrinadina A, citrinadina A, chrisogenamida A e triptoquialaninas são produzidas
na interface entre os dois microrganismos. A ação antifúngica dos metabólitos
envolvidos na interação foi confirmada através de ensaios antimicrobianos. As
triptoquialaninas inibiram a formação de esporos de P. citrinum, enquanto a
crisogenamida A, citrinadinas e tetrapeptídeos inibiram o crescimento radial de P.
digitatum. Este trabalho demonstra que a estratégia de co-cultivo em conjunto com
MSI é útil no descobrimento de novos compostos bioativos e no entendimento da
ecologia microbiana.
 www.nature.com/scie
114

6.2 Artigo: “Antifungal potential of secondary metabolites involved in the interaction between citrus
pathogens”

OPEN Antifungal potential of secondary

metabolites involved in the
interaction between citrus
pathogens
Jonas Henrique Costa1,5, Cristiane Izumi Wassano1,5, Célio Fernando Figueiredo Angolini 2
,
Kirstin Scherlach3, Christian Hertweck 3,4 & Taícia Pacheco Fill1*

Numerous postharvest diseases have been reported that cause substantial losses of citrus fruits
worldwide. Penicillium digitatum is responsible for up to 90% of production losses, and represent a
problem for worldwide economy. In order to control phytopathogens, chemical fungicides have been
extensively used. Yet, the use of some artificial fungicides cause concerns about environmental risks and
fungal resistance. Therefore, studies focusing on new approaches, such as the use of natural products,
are getting attention. Co-culture strategy can be applied to discover new bioactive compounds and
to understand microbial ecology. Mass Spectrometry Imaging (MSI) was used to screen for potential
antifungal metabolites involved in the interaction between Penicillium digitatum and Penicillium
citrinum. MSI revealed a chemical warfare between the fungi: two tetrapeptides, deoxycitrinadin
A, citrinadin A, chrysogenamide A and tryptoquialanines are produced in the fungi confrontation
zone. Antimicrobial assays confirmed the antifungal activity of the investigated metabolites. Also,
tryptoquialanines inhibited sporulation of P. citrinum. The fungal metabolites reported here were never
described as antimicrobials until this date, demonstrating that co-cultures involving phytopathogens
that compete for the same host is a positive strategy to discover new antifungal agents. However, the
use of these natural products on the environment, as a safer strategy, needs further investigation. This
paper aimed to contribute to the protection of agriculture, considering health and ecological risks.

Citrus fruits have an important impact on world’s economy since they have the largest production compared with
other fruits1; for example, the global orange production for 2018/19 is forecast to reach 54.3 million tons2. The
major factor affecting the quality of citrus is the postharvest fungal diseases, in particular the green mold caused
by Penicillium digitatum that is responsible for 90% of citrus losses during postharvest period1,3.
To control post-harvest diseases, synthetic fungicides are widely used4, causing health and environmental
issues5. Furthermore, some fungi strains have developed resistance to commonly used fungicides6. As antifungal
resistance is becoming a significant concern, the search for new bioactive compounds has an important role to
bypass the extensive use of fungicides7.
Microorganisms are one of the main sources for natural products with useful biological activities, i.e., poten-
tial antifungals, antibiotics, anticancer agents, surfactants8,9. However, besides the genetic potential and diversity
of microbes, many microbial biosynthetic genes are not activated in the unnatural cultivation conditions used in
laboratory and only a few part of the metabolites are accessible10. Several strategies are implemented to overcome
the limitations in natural products discovery from microbial sources, i.e., OSMAC approach10,11, the use of epige-
netic modifications12,13 and co-cultivation14.
Co-cultures that mimic natural environments, exhibiting microbial competition for limited space and nutri-
ents, have revealed to be a major ecological force that could activate silent gene clusters and defense mecha-
nisms that can lead to the production of bioactive secondary metabolites14–17. Co-culture experiments are highly

1
Institute of Chemistry, University of Campinas, CP 6154, 13083-970, Campinas, SP, Brazil. 2Center for Natural and
Human Sciences, Federal University of ABC, 09210-580, Santo André, SP, Brazil. 3Department of Biomolecular
Chemistry, Leibniz Institute for Natural Product Research and Infection Biology – Hans Knöll Institute, Jena,
Germany. 4Chair of Natural Product Chemistry, Friedrich Schiller University Jena, 07743, Jena, Germany. 5These
authors contributed equally: Jonas Henrique Costa and Cristiane Izumi Wassano. *email: taicia@unicamp.br

SCIENTIFIC REPORTS | (2019) 9:18647 | https://doi.org/10.1038/s41598-019-55204-9 1

www.nature.com/scientificreports/ www.nature.com/scientificreports
115

relevant for allowing not only the identification of new compounds, but also to investigate chemical events that
govern interactions between microorganisms in nature18–21.
To date, there is no information about how citrus pathogen fungi interact with other microorganisms in the
same environment and what are the mechanisms of attack and defense against endophytic microorganisms or
other phytopathogens. This information may lead to the discovery of new secondary metabolites involved in
microbial interactions and provide better knowledge about the importance of microbial ecology during infection
process.
Here we show an interaction between Penicillium digitatum and Penicillium citrinum, aiming to search for new
antifungal compounds that could be used to control postharvest diseases. Considering this purpose, co-culture
strategy and MSI were applied to induce the production of secondary metabolites and provide initial insights
concerning their biological role based on their spatial distribution in the interaction. The metabolites of interest
were isolated and all the structures were elucidated based on Mass spectrometry and NMR experiments. Yet,
antifungal assays and confocal microscopy analyses were performed in order to investigate the potential of the
fungal metabolites as new antifungal agents.

Materials and Methods

Fungi culture. The P. digitatum (PD) strain used in the studies is deposited with the Spanish Type Culture
Collection (CECT) under the accession code CECT20796. P. citrinum (PC) strain was provided by Sylvio Moreira
Citrus Center (Cordeirópolis, SP, Brazil). P. digitatum and P. citrinum were cultivated on commercial potato dex-
trose agar (PDA) (Acumedia). PDA was autoclaved at 103 KPa (121 °C) for 15 min. PDA plates were stored at
25 °C for 7 days in darkness. Spores were harvested by washing the agar surface with sterile distilled water and
diluted to a final concentration of 106 or 105 spore mL−1.

Co-culture growth conditions. In vitro co-culture was prepared in 20 mL of PDA plates and 5 µL of each
fungal spore solution (106 spore mL−1) was inoculated, on opposite sides. The plates were incubated in darkness
at 25 °C for 7 days.
For MSI analyses and confocal microscopy, the in vitro co-culture was prepared by placing a sterile micro-
scope slide in the Petri dish, followed by the pouring of 11 ml of PDA22 and the inoculums were made above the
microscope slide. The plates were incubated in darkness at 25 °C for 72 h.
For in vivo co-culture, mature oranges (Citrus sinensis) were surface sterilized and wounded23. A small piece of
PDA containing P. citrinum was inoculated in the wound site. Infected and control oranges were stored in sterile
500 mL beakers, in darkness at 25 °C. After 10 days, the fruits were wounded on the opposite equatorial region of
P. citrinum inoculum and infected with P. digitatum 106 spore mL−1 solution. The fruits were stored for more 5
days in darkness at 25 °C.

Mass Spectrometry Imaging (MSI) analysis and MS image generation. After the incubation
period, the microscope slides were removed from the Petri dishes and put in a vacuum desiccator, for 1 hour,
for complete agar dehydration (Angolini et al., 2015). MSI analyses were performed directly on the microscope
containing the co-culture, in positive mode, using a desorption electrospray ionization (DESI) source Prosolia
® ®
Model Omni Spray 2D -3201) coupled to a Thermo Scientific QExactive Hybrid Quadrupole-Orbitrap Mass
Spectrometer. MSI data was acquired with a mass resolving power of 70.000 at m/z 200. The DESI configuration
used was the same set by previous work22. Images were generated with a bin width of ∆m/z = ± 0.07 using
Firefly data conversion software (version 2.1.05) and processed using BioMap software (version 3.8.0.4) devel-
oped by Novartis Institutes for BioMedical Research. In BioMap, color scaling was adjusted to a fixed value during
the processing of each image. MS spectra were processed with Xcalibur software (version 3.0.63) developed by
Thermo Fisher Scientific.

Extraction of metabolites from the co-culture experiments. The whole contents of the co-culture
in vitro were cut into small pieces and transferred to an Erlenmeyer flask. The extraction was performed using
methanol. The flasks were sonicated during 1 h in ultrasonic bath and vacuum filtered. Solvent was removed
under reduced pressure and the final extract stored at −20 °C.
For in vivo co-culture, the orange peels were cut (2 cm × 2 cm) in the interface zone between the microorgan-
isms and extraction was performed with 5 mL of methanol during 1 h in ultrasonic bath. Extracts were filtered,
dried under N2 and stored at −20 °C.

Mass Spectrometry analysis (MS). Extracts were diluted in methanol and analyzed on a Thermo Scientific
®
QExactive Hybrid Quadrupole-Orbitrap Mass Spectrometer. Analyses were performed in the positive mode
with m/z range of 115–1500, capillary voltage of 3.4 kV, inlet capillary temperature of 280 °C, S-lens 100 V. 5 µL
of sample were injected. Stationary phase: Thermo Scientific column Accucore C18 2.6 µm (2.1 mm × 100 mm).
Mobile phase: 0.1% formic acid (A) and acetonitrile (B). Eluent profile (A/B): 95/5 up to 2/98 within 15 min, hold
for 5 min, up to 95/5 within 1.2 min and hold for 7.8 min. The total run time was 29 min for each run and the flow
rate, 0.2 mL min−1. Injection volume: 5 µL.
MS/MS was performed by the collision induced dissociation (CID) with m/z range of 100–800 and the col-
lision energy ranged from 10 to 50 V. The samples were directly infused by electrospray with 5.0 µL min−1 flow
rate. MS and MS/MS data was processed with Xcalibur software (version 3.0.63) developed by Thermo Fisher
Scientific.

Metabolite separation (HPLC analysis). Secondary metabolites separations were achieved using a
Phenomenex column Luna 5 µm Phenyl-Hexyl (250 × 4.6 mm) and a SHIMADZU prominence HPLC LC-20AT,
equipped with CBM-20A communication bus module, SPD-M20A photodiode array detector and SIL-20A auto

SCIENTIFIC REPORTS | (2019) 9:18647 | https://doi.org/10.1038/s41598-019-55204-9 2

www.nature.com/scientificreports/ www.nature.com/scientificreports
116

sampler. Mobile phase: water (0.1% formic acid) (A) and acetonitrile (B). Eluent profile (A/B): 65/35 up to 50/50
within 50 min, up to 40/60 within 20 min. The total run time was 70 min and the flow rate of 1.0 mL min−1.
Injection volume was 5 µL. Preparative HPLC purifications were performed on a Phenomenex column Luna 5 µm
Phenyl-Hexyl (250 × 10 mm) using a Waters 1525 Binary HPLC Pump equipped with Waters 2998 Photodiode
Array Detector and Waters Fraction Collector III using the same optimized gradient conditions. The flow rate was
set at 4.7 mL min−1 and the injection volume was 200 µL.

NMR Spectroscopy. 1H NMR, 13C NMR and 2D experiments were performed on a Bruker Avance III
500 (1H 500.13 MHz and 13C 125.7 MHz) and Bruker Avance III 600 (1H 600.17 MHz). Deuterated chloroform
(CDCl3; 7.23 ppm), dimethyl sulfoxide (DMSO; 2.50 ppm and 39.51 ppm) and tetramethylsilane (TMS; 0.0 ppm)
were used as a solvent and internal reference. Chemical shifts (δ) were expressed in (ppm) and the coupling con-
stants (J) in Hertz (Hz).

Molecular Networking analysis. A molecular network for P. citrinum metabolites was created using the
online workflow at Global Natural Products Social Molecular Networking (GNPS) (http://gnps.ucsd.edu). The
data was filtered by removing all MS/MS peaks within + /−17 Da of the precursor m/z. MS/MS spectra were
window filtered by choosing only the top 6 peaks in the + /−50 Da window throughout the spectrum. The data
was then clustered with MS-Cluster with a parent mass tolerance of 0.2 Da and a MS/MS fragment ion toler-
ance of 0.2 Da to create consensus spectra. Further, consensus spectra that contained less than 2 spectra were
discarded. A network was then created where edges were filtered to have a cosine score above 0.65 and more
than 2 matched peaks. Further edges between two nodes were kept in the network only if each of the nodes
appeared in each other’s respective top 10 most similar nodes. The spectra in the network were then searched
against GNPS’ spectral libraries. The library spectra were filtered in the same manner as the input data. All
matches kept between network spectra and library spectra were required to have a score above 0.5 and at least
5 matched peaks. The resulting molecular networking is available at https://gnps.ucsd.edu/ProteoSAFe/status.
jsp?task=c6f716c6fd044ba985eacd96935ee0c3.

Antifungal assays. A stock solution of the co-culture extract was prepared in methanol and further diluted
in PDA to the concentration of 0.5 mg mL−1. 15 mL of the resultant solution was poured in a Petri dish followed
by the inoculation of 15 µl of a 105 spore mL−1 P. digitatum solution on the center of the agar plate. A control assay
was also performed. The plates were incubated in darkness at 25 °C for 96 h.
For antifungal assays, 2.5 mL of PDA was supplemented with 6, 9 and 10 (400 µg mL−1) and each solution were
poured in a 6-well microplate. 5 µl of a 105 spore mL−1 P. digitatum solution was inoculated on the center of each
agar plate. Negative controls were performed in triplicate. The microplate was incubated in darkness at 25 °C for
96 h.
For determination of minimum inhibitory concentration (MIC) of compounds 1 and 4, microbroth dilution
assay was performed as recommended by Clinical and Laboratory Standards Institute (2008)24 with few modi-
fications. Stock solutions of 1 and 4, were prepared in water (5% methanol) and further diluted in YES media
in a range of concentrations to 600 µg mL−1 to 1 µg mL−1. 195 µl of each solution were transferred to a 96-well
microplate followed by the inoculation of 5 µl of a 105 spore mL−1 P. digitatum solution. Assays were made in
duplicate and controls in triplicate. Itraconazole (100 µg mL−1) were used as positive control. Negative controls
were performed with methanol in YES. The microplates were incubated in darkness at 25 °C for 96 h.

Confocal laser scanning microscopy. After the incubation period, the microscopes slides were removed
from the Petri dish. The in vitro co-culture samples were stained with Congo Red (0.25% w/v in water) for 20 min-
utes and briefly washed in distilled water. Samples were analyzed with Leica TCS SP5 microscope. Excitation was
by the 543 nm emission line of the He-Ne laser, and light emitted between 570 and 680 nm was collected25.

Results and Discussion

MSI reveals potential antifungals in the interaction between P. digitatum and P. citrinum. To
screen for new antifungal compounds with potential to protect citrus fruits and control postharvest diseases,
we applied a co-culture strategy involving P. digitatum and another citrus pathogen, P. citrinum. Co-culture is
a strategy inspired by nature in which the competition between the microorganisms can induce the production
of new metabolites8. In previous work, the co-cultivation between Trichophyton rubrum and Bionectria ochro-
leuca induced the production of a new sulfated analogue of PS-990, suggesting that this compound is further
sulfated during the fungal interaction26. Also, another example of a compound derived from fungi interaction is
the tetrapeptide cyclo-(L-leucyl-trans-4-hydroxy-L-prolyl-D-leucyl-trans-4-hydroxy-L-proline) isolated from the
co-culture broth of Phomopsis sp. K38 and Alternaria sp. E33; the cyclic tetrapeptide exhibited moderate to high
inhibitory activity against phytopathogenic fungi when compared to the commercial fungicide triadimefon27.
Thus, co-cultivation experiments are a viable approach to find compounds that can inhibit the main citrus phyto-
pathogens, specially, the green mold caused by P. digitatum.
In co-cultivation performed in both orange (in vivo) and synthetic media (in vitro) we visually observed a
long-distance growth inhibition between P. citrinum and P. digitatum. In a fungi interaction, silent genes can be
activated and harmful metabolites can be diffused from one partner to the other16,28. These induced metabolites
are usually localized at the zone of confrontation in solid media of co-cultures16. However, regular approaches
used to detect and elucidate metabolites such as mass spectrometry coupled to liquid (LC-MS) or gas (GC-MS)
chromatography do not provide information about the spatial distribution of the molecules29.
The information about molecular spatial distribution can be obtained by mass spectrometry imaging (MSI), a
powerful tool that generates images for each ion detected in the mass spectrum16. The use of MSI to understand
microbial systems and their secondary metabolites is not new and studies where this technique was successfully

SCIENTIFIC REPORTS | (2019) 9:18647 | https://doi.org/10.1038/s41598-019-55204-9 3

www.nature.com/scientificreports/ www.nature.com/scientificreports
117

Figure 1. Imaging analysis by (+) DESI-MS of citrus pathogens co-culture. Each MS image shows the spatial
distribution of a m/z ratio over the co-culture surface. All images are plotted on the same color scale from 0
(black) to 2 × 105 (red) yet ion concentration cannot be compared across images since differences in chemical
structures can lead to variation in ionization efficiency30.

applied can be found in the literature30. By example, MSI was applied to investigate the interaction between
Bacillus subtilis 3610 and Streptomyces coelicolor A3. DESI imaging of the bacterial co-culture revealed 57 signals
spatially localized to bacterial colonies, leading to the identification of some secondary metabolites such as sur-
factin and plipastatin31. Furthermore, MSI analysis also showed that S. coelicolor has the production of certain
secondary metabolites inhibited in the presence of B. subtilis, revealing an interaction between bacteria31.
Therefore, to detect the secondary metabolites involved in the interaction between the citrus pathogens, we
applied DESI-MSI directly on the surface of the co-culture agar, to visualize the diffusion of compounds to the
zone of confrontation. MSI signals were obtained for ions [M + H]+ at m/z 519.1857, m/z 503.1908, m/z 475.1590,
m/z 460.1960, m/z 459.1645, m/z 625.3942, m/z 609.3988, m/z 527.2862, m/z 511.2894 and m/z 448.2938
(Figs. S1–S10). We observed that all the ions mentioned were detected and concentrated in the zone of confron-
tation between the fungi (Fig. 1). These compounds may be related to the fungus-fungus interaction and could be
potentially new antimicrobial agents. Ions at m/z 625, 609, 527, 511 and 448 were produced by P. citrinum, while
ions at m/z 519, 503, 475, 460 and 459, seemed to be a counter-attack of P. digitatum, revealing a chemical warfare
between these two citrus pathogens. The ions detected in vitro through MSI analyses were also detected in the
extracts of the co-cultures in vivo (Figs. S11–S20) using oranges as substrate (Fig. 2).
To characterize the metabolites involved in the interaction, a co-culture extract was obtained from a scale up
cultivation experiment and the compounds of interest, detected initially through MSI analyses, were isolated by
preparative HPLC and characterized through tandem mass spectrometry and NMR analyses.

SCIENTIFIC REPORTS | (2019) 9:18647 | https://doi.org/10.1038/s41598-019-55204-9 4

www.nature.com/scientificreports/ www.nature.com/scie
118

Figure 2. Overview of the experimental setup and strategies applied to analyze the secondary metabolites
involved in citrus pathogens interaction. P. digitatum and P. citrinum were co-cultured in orange (in vivo) and
in PDA media (in vitro) to induce the production of secondary metabolites. MSI was applied, in vitro, to detect
the secondary metabolites produced in the zone of confrontation. HRMS analysis was performed to confirm
the fungus-fungus interaction in vivo. The metabolites of interest were isolated from a scale up experiment.
Subsequently, antifungal assays and confocal laser scanning microscopy analysis were performed to investigate
antimicrobial activity and fungal cell morphology, respectively.

Ion formula Calculated Experimental Error

Compound ([M + H]+) m/z m/z (ppm)
Tryptoquialanine A C27H27N4O7 519.1874 519.1857 −3.4
Tryptoquialanine C C27H27N4O6 503.1925 503.1908 −3.5
Tryptoquialanone C25H23N4O6 475.1612 475.1590 −4.6
15-dimethyl-2-epi-fumiquinazoline A C25H26N5O4 460.1979 460.1960 −4.2
deoxytryptoquialanone C25H23N4O5 459.1663 459.1645 −4.0
Citrinadin A C35H53N4O6 625.3960 625.3942 −2.8
Deoxycitrinadin A C35H53N4O5 609.4010 609.3988 −3.7
Phe-Val-Val-Tyr C28H39N4O6 527.2864 527.2862 −0.4
Phe-Val-Val-Phe C28H39N4O5 511.2915 511.2894 −4.1
Chrysogenamide A C28H38N3O2 448.2959 448.2938 −4.6

Table 1. DESI-MSI data obtained for the secondary metabolites involved in P. digitatum and P. citrinum
interaction.

Through the exact masses obtained by DESI-MSI analyses (Table 1) it was possible to confirm the presence
of indole alkaloids produced by P. digitatum. Ions [M + H]+ at m/z 519.1857, m/z 503.1908, m/z 475.1590, m/z
460.1960 and m/z 459.1645 which correspond respectively to tryptoquialanine A (1), tryptoquialanine C (2),
tryptoquialanone (3), 15-dimethyl-2-epi-fumiquinazoline A (4) and deoxytryptoquialanone (5) (chemical struc-
tures represented in Fig. 3). These compounds are part of the tryptoquialanines biosynthetic pathway32 and were
previously detected during DESI-MSI analysis of oranges infected with the green mold disease23.
Compound 1 was reported as the major secondary metabolite produced by P. digitatum33. Yet, the deletion
of tqaA gene, responsible to regulate 1 biosynthesis, showed that 1 is not involved in the pathogenicity of P. dig-
itatum against citrus since the infection ability of the mutants was not altered34. The exact biological role of the
tryptoquialanines is still unknown34 and this study provides a new biological activity of the tryptoquialanines
with the involvement of these alkaloids in the fungal-fungal interaction.
The MS/MS of the ion [M + H]+ at m/z 625.3951 yielded to fragments at m/z 594.3533, 576.3427, 449.2430
and 431.2325 (Fig. S21), the same fragmentation pattern obtained for citrinadin A (6) in a study involving
co-culture between P. citrinum and Pseudoalteromonas sp. OT5919. Also GNPS database suggested that the ion
at m/z 625 could be citrinadin A. The tandem mass spectrum obtained for this compound shared five fragments

SCIENTIFIC REPORTS | (2019) 9:18647 | https://doi.org/10.1038/s41598-019-55204-9 5

www.nature.com/scientificreports/ www.nature.com/scie
119

Figure 3. Chemical structures of indole alkaloids produced by P. digitatum: tryptoquialanine A

(1), tryptoquialanine C (2), tryptoquialanone (3), 15-dimethyl-2-epi-fumiquinazoline A (4) and
deoxytryptoquialanone (5). These compounds were detected through DESI-MSI in the fungal confrontation
zone.

in common with the tandem mass spectrum of 6 deposited in the database (Fig. S22). In addition, it shares sim-
ilar MS/MS fragmentation pattern with citrinadin A reported by Moree et al. (2014). Citrinadins were reported
being higher produced in co-culture situations and concentrated in co-culture interfaces, suggesting a defensive
response of P. citrinum against other microorganisms19,35.
Molecular networking analysis of the isolated compounds revealed that another citrinadin is involved in the
co-culture interaction. We observed that ion at m/z 609, detected initially by MSI analysis, is grouped in the same
cluster of compound 6 (m/z 625) (Fig. S23), suggesting that this compound is a citrinadin-like metabolite. In a
molecular networking analysis, related metabolites are grouped in same clusters since they have similar MS/MS
spectrum36.
The exact mass obtained for ion [M + H]+ at m/z 609.4010 corresponds to a compound with molecular for-
mula C35H53N4O5. Fragmentation pattern yielded to fragments at m/z 578.3583, 464.2905, 451.2586 and 433.2482
(Supplementary Fig. S24). 1H NMR and 1H-1H COSY spectra obtained for the purified metabolite (Figs. S25–S26
and Table S1) exhibited similar signals of a synthesized deoxycitrinadin A that was reported by Bian et al. (2013)37.
The epoxide characteristic signal at δH 4.0 was absent in this derivative and a signal for vynilic hydrogen could be
observed at δH 6.9, indicating the lack of the epoxide group and the presence of a double bond in comparison with
citrinadin A37. It is the first time that deoxycitrinadin A (7) is reported in the literature as a secondary metabolite
produced by a microorganism.
Exact masses obtained by DESI-MSI for ions [M + H]+ at m/z 527.2862 and 511.2894 indicated compounds
with elemental composition of C28H38N4O6 and C28H38N4O5, respectively. MS/MS revealed that the ion at m/z
511 has similar structure compared to m/z 527, except by an absence of an oxygen atom. Fragmentation of the
ion at m/z 527 yielded to fragments m/z 281, 247, 219, 182 and 120 (Fig. S27), while ion at m/z 511 yielded to m/z
265, 247, 219, 166 and 120 (Fig. S30). Same fragmentation pattern was reported for the sequence of tetrapeptides
Phe-Val-Val-Tyr (8) of Penicillium canescens38. Yet, the tetrapeptides 8 and Phe-Val-Val-Phe (9) were recently
reported as secondary metabolites produced by Penicillium roqueforti39. 1H and 13C NMR analyses of the isolated
compounds (Figs. S28–S32 and Tables S2-S3) confirmed the results obtained through MS/MS, concluding that
ions at m/z 527 and m/z 511 correspond to 8 and 9, respectively. The production of these tetrapeptides in fungal
chemical warfare is not surprising because small peptides are known for their antimicrobial activity40. This is the
first report of 8 and 9 as secondary metabolites of P. citrinum.
For ion [M + H]+ at m/z 448.2938, the exact mass suggested a compound with molecular formula C28H38N3O2,
the same composition of the secondary metabolite chrysogenamide A (10) (error = −4.6 ppm). Compound 10
was isolated and 1H and 13C NMR analyses (Figs. S33–S34 and Table S4) confirmed its structure; it is the first
report that chrysogenamide A is involved in a fungal-fungal interaction. Compound 10 was first reported as
a secondary metabolite of Penicillium chrysogenum No. 005, an endophytic fungus associated with the plant
Cistanche deserticola41. Also, 10 was reported as a secondary metabolite of a P. citrinum strain42. The structures of
the metabolites produced by P. citrinum are represented in Fig. 4.

SCIENTIFIC REPORTS | (2019) 9:18647 | https://doi.org/10.1038/s41598-019-55204-9 6

www.nature.com/scientificreports/ www.nature.com/scie
120

Figure 4. Chemical structures of secondary metabolites produced by P. citrinum during chemical warfare
against P. digitatum: citrinadin A (6), deoxycitrinadin A (7) and chrysogenamide A (10).

Antifungal assays. We investigated the antifungal activity of the metabolites produced in the co-culture and
P. digitatum was inoculated in agar media containing co-culture extract. Compared to the control, we observed
a reduction of 67% in P. digitatum radial growth (Fig. S35), indicating that the metabolites involved in the fungal
warfare have potential as antifungal agents. To confirm this hypothesis, we tested the isolated compounds and
observed that compounds 6, 9 and 10 (400 µg mL−1) reduced 48%, 41% and 61% of P. digitatum radial growth,
respectively, when compared to control (Fig. 5), confirming the antifungal activity.
Citrinadins were found to be involved in the response of P. citrinum against other microorganisms in
co-cultures, but the biological role of them in biological environments are still unknown to this date19. Compound
6 was tested for Anti-buruli ulcer activity on Mycobacterium ulcerans MN209, but no interesting MIC was
observed43; also, cytotoxicity activity of 6 against leukemia and carcinoma cells was reported44. Our results are the
first report of an antimicrobial activity of citrinadins in literature and can provide first insights about the biologi-
cal role of these compounds in fungal-fungal interactions.
In addition, chrysogenamide A was never reported as an antimicrobial agent until now. Compound 10 exhib-
ited a protective effect on neurocytes against oxidative stress-induced cell death41, however no other biological
activity for 10 was reported in the literature.
Antifungal assays applying D-Phe-L-Val-D-Val-L-Tyr revealed that this tetrapeptide has inhibitory activity
against B. subitllis and the soybean phytopathogen Fusarium virguliforme38. In contrast, no inhibition of E. coli,
B. subtilis and S. cerevisiae in presence of D-Phe-L-Val-D-Val-L-Tyr or D-Phe-L-Val-D-Val-L-Phe was observed39.
The antifungal activity of the tryptoquialanines was also evaluated, since these compounds seemed to be a
chemical response of P. digitatum against P. citrinum in the chemical warfare. Tryptoquialanines 1 and 4 were
tested and revealed an antifungal activity against P. citrinum. Compounds 1 and 4 had MIC of 300 µg mL−1, inhib-
iting P. citrinum spore production (Fig. S36). To the best of our knowledge, it is the first report of an antimicrobial
activity of the tryptoquialanines. Recently, 1 was demonstrated as an insecticidal compound against Ae. aegypti
larvae23; the antifungal activity can provide more understanding about the role of the tryptoquialanines in the
citrus-pathogen environment once these compounds are not required for P. digitatum virulence34.

Chemical warfare alters P. digitatum cell wall in co-culture. To investigate the action of the second-
ary metabolites during the fungal interaction, co-culture samples were stained with Congo Red and observed
through confocal laser scanning microscopy. Congo Red is commonly used to stain polysaccharides containing
β 1,4 linkages as, by example, the fungal cell wall component chitin45,46. P. digitatum hyphae were observed in
the confrontation zone (sample) and compared with hyphae distant to the interface region (control) (Fig. S37).
Control hyphae were homogeneous stained with Congo Red while hyphae in the interface region exhibited an
altered staining pattern (Fig. 6), with irregular patches. Similar staining patterns were obtained for knockout

SCIENTIFIC REPORTS | (2019) 9:18647 | https://doi.org/10.1038/s41598-019-55204-9 7

www.nature.com/scientificreports/ www.nature.com/scientificreports
121

Figure 5. P. digitatum growing on PDA with 400 µg ml−1 of (A) chrysogenamide A (B) citrinadin A and (C)
tetrapeptide Phe-Val-Val-Phe. In 96 h, an inhibition in radial growth is observed in the presence of the isolated
metabolites when compared to (D) control.

Figure 6. Confocal laser scanning microscopy of Congo Red-stained P. digitatum hyphae (A) distant from
P. citrinum and (B) in the zone of confrontation. Patches of Congo Red indicates a defective fungal cell wall.
Bars = 5.0 µm.

mutant fungi which the deleted genes had a role in cell wall organization; as result, mutants exhibited defective
cell walls and irregular staining25,46,47. Yet, abnormal staining with Calcofluor White was observed for P. ostreatus
P89 treated with 36 °C; high temperature altered chitin distribution and cell wall integrity48.
This data shows that P. digitatum hyphae, in contact with the metabolites diffused during the co-culture, have
a defective cell wall since Congo Red bounds to fungal cell wall structures. The fungal cell wall is an attractive
target of antimicrobials because they are not present in mammalian cells49,50. In conclusion, the microscopy anal-
ysis and the antifungal assays reinforce that the metabolites involved in the fungal interaction have potential as
antifungal agents and may be the mechanism in nature that these phytopathogens developed to compete against
other microorganisms for the host (Fig. 7).

Conclusions
The search for new natural antimicrobials is a promising field in natural products research concerning the eco-
nomic impact of postharvest diseases to worldwide agriculture. Furthermore, the appearance of fungi strains
resistant to fungicides makes the discovery of new antifungal agents to replace synthetic compounds extremely
important. Using co-cultivation between phytopathogens that compete for the same host, P. digitatum and P. cit-
rinum, we observed a fungal interaction. Through MSI technique, we detected secondary metabolites diffused to
the interface zone between the microorganisms. Tryptoquialanines, citrinadins, chyrsogenamide A and tetrapep-
tides exhibited great antifungal activity, confirming that co-cultures and MSI technique are a good combination
in the search of new natural antimicrobials.
Until this date, there has been no information about the interaction between citrus pathogenic fungi. Our
data revealed compounds that play a role in the citrus microbial ecology. In addition, we demonstrated that the
metabolites studied have great potential as antifungal agents since fungal cell walls are one of the main targets of
antifungal compounds. The use of the identified compounds as natural antifungals instead of synthetic fungicides
should be further investigated. This paper opens new research possibilities and contributes to the environmental

SCIENTIFIC REPORTS | (2019) 9:18647 | https://doi.org/10.1038/s41598-019-55204-9 8

www.nature.com/scientificreports/ www.nature.com/scientificreports
122

Figure 7. Chemical warfare between P. citrinum and P. digitatum in citrus fruit. Tryptoquialanines, citrinadins,
chrysogenamide A, tetrapeptides and other metabolites are involved in the long-distance inhibition observed.

and human health, helping in the search of safer strategies for agriculture through the use of compounds obtained
from natural sources.

Data availability
All data generated or analyzed during this study are included in this published article and its Supplementary
Information file.

Received: 17 September 2019; Accepted: 23 November 2019;

Published: xx xx xxxx

References
1. Ghooshkhaneh, N. G., Golzarian, M. R. & Mamarabadi, M. Detection and classification of citrus green mold caused by Penicillium
digitatum using multispectral imaging. J Sci Food Agric 98(9), 3542–3550, https://doi.org/10.1002/jsfa.8865 (2018).
2. United States Department of Agriculture - Foreign Agricultural Service, 2019. Citrus: World Markets and Trade, https://apps.fas.
usda.gov/psdonline/circulars/citrus.pdf (accessed 06 August 2019).
3. Macarisin, D. et al. Penicillium digitatum suppresses production of hydrogen peroxide in host tissue during infection of citrus fruit.
Phytopathology 97, 1491–1500, https://doi.org/10.1094/PHYTO-97-11-1491 (2007).
4. Hao, W., Li, H., Hu, M., Yang, L. & Rizwan-ul-Haq, M. Integrated control of citrus green and blue mold and sour rot by Bacillus
amyloliquefaciens in combination with tea saponin. Postharvest Biol Technol 59, 316–323, https://doi.org/10.1016/j.
postharvbio.2010.10.002 (2011).
5. Frisvad, J. C. & Samson, R. A. Polyphasic taxonomy of Penicillium subgenus Penicillium A guide to identification of food and air-
borne terverticillate Penicillia and their mycotoxins. Stud Mycol 49, 1–174 (2004).
6. Kanetis, L., Förster, H. & Adaskaveg, J. E. Determination of natural resistance frequencies in Penicillium digitatum using a new air-
sampling method and characterization of Fludioxonil- and Pyrimethanil-Resistant isolates. Phytopathology 100, 738–743, https://
doi.org/10.1094/PHYTO-100-8-0738 (2010).
7. Strano, M. C., Altieri, G., Admane, N., Genovese, F. & Di Renzo, G. C. Advance in Citrus Postharvest Management: Diseases, Cold
Storage and Quality Evaluation. In: Gill, H. & Garg, H. Citrus pathology. Ch 7,139–159, https://doi.org/10.5772/66518 (2017).
8. Azzollini, A. et al. Dynamics of Metabolite Induction in Fungal Co-cultures by Metabolomics at Both Volatile and Non-volatile
Levels. Front Microbiol 9, 72, https://doi.org/10.3389/fmicb.2018.00072 (2018).
9. Demain, A. L. Importance of microbial natural products and the need to revitalize their discovery. J Ind Microbiol Biotechnol 41,
185–201, https://doi.org/10.1007/s10295-013-1325-z (2014).
10. Bode, H. B., Bethe, B., Höfs, R. & Zeeck, A. Big effects from small changes: Possible ways to explore nature’s chemical diversity.
ChemBioChem, 3, 619–627, 10.1002/1439-7633(20020703)3:7<619::AID-CBIC619>3.0.CO;2-9 (2002).
11. Pan, R., Bai, X., Chen, J., Zhang, H. & Wang, H. Exploring Structural Diversity of Microbe Secondary Metabolites Using OSMAC
Strategy: A Literature Review. Front Microbiol 10, 294, https://doi.org/10.3389/fmicb.2019.00294 (2019).
12. Keller, N. P. Fungal secondary metabolites: regulation, function and drug discovery. Nat Rev Microbiol 17, 167–180, https://doi.
org/10.1038/s41579-018-0121-1 (2019).
13. Aghcheh, R. K. & Kubicek, C. P. Epigenetics as an emerging tool for improvement of fungal strains used in biotechnology. Appl
Microbiol Biotechnol 99(15), 6167–6181, https://doi.org/10.1007/s00253-015-6763-2 (2015).
14. Marmann, A., Aly, A. H., Lin, W., Wang, B. & Proksch, P. Co-Cultivation – A Powerful Emerging Tool for Enhancing the Chemical
Diversity of Microorganisms. Mar Drugs 12, 1043–1065, https://doi.org/10.3390/md12021043 (2014).
15. Netzker, T. et al. Microbial communication leading to the activation of silent fungal secondary metabolite gene clusters. Front
Microbiol 6, 299, https://doi.org/10.3389/fmicb.2015.00299 (2015).
16. Bertrand, S. et al. Metabolite induction via microorganism co-culture: A potential way to enhance chemical diversity for drug
discovery. Biotechnol Adv 32(6), 1180–1204, https://doi.org/10.1016/j.biotechadv.2014.03.001 (2014).
17. Vinale, F. et al. Co-Culture of Plant Beneficial Microbes as Source of Bioactive Metabolites. Sci. Rep. 7, 14330, https://doi.
org/10.1038/s41598-017-14569-5 (2017).
18. Chagas, F. O., Dias, L. G. & Pupo, M. T. A mixed culture of endophytic fungi increases production of antifungal polyketides. J Chem
Ecol 39(10), 1335–1342, https://doi.org/10.1007/s10886-013-0351-7 (2013).

SCIENTIFIC REPORTS | (2019) 9:18647 | https://doi.org/10.1038/s41598-019-55204-9 9

www.nature.com/scientificreports/ www.nature.com/scientificreports
123

19. Moree, W. J. et al. Microbiota of Healthy Corals Are Active against Fungi in a Light-Dependent Manner. ACS Chem Biol 9(10),
2300–2308, https://doi.org/10.1021/cb500432j (2014).
20. Partida-Martinez, L. P. & Hertweck, C. Pathogenic fungus harbous endosymbiotic bacteria for toxin production. Nature 437,
884–888, https://doi.org/10.1038/nature03997 (2005).
21. Heine, D. et al. Chemical warfare between leafcutter ant symbionts and a co-evolved pathogen. Nat. Commun. 9, 2208, https://doi.
org/10.1038/s41467-018-04520-1 (2018).
22. Angolini, C. F. F. et al. Direct Protocol for Ambient Mass Spectrometry Imaging on Agar Culture. Anal Chem 87, 6925–6930, https://
doi.org/10.1021/acs.analchem.5b01538 (2015).
23. Costa, J. H. et al. Monitoring indole alkaloid production by Penicillium digitatum during infection process in citrus by Mass
Spectrometry Imaging and molecular networking. Fungal Biol 123(8), 594–600, https://doi.org/10.1016/j.funbio.2019.03.002
(2019).
24. CLSI. Reference Method for Broth Dilution Antifungal Susceptibility Testing of Filamentous Fungi; Approved Standard – Second
Edition. CLSI document M38-A2. Wayne, PA: Clinical and Laboratory Standards Institute; 2008.
25. Siriputthaiwan, P., Jauneau, A., Herbert, C., Garcin, D. & Dumas, B. Functional analysis of CLPT1, a Rab/GTPase required for
protein secretion and pathogenesis in the plant fungal pathogen Colletotrichum lindemuthianum. J Cell Sci 118, 323–329, https://
doi.org/10.1242/jcs.01616 (2005).
26. Bertrand, S. et al. De Novo Production of Metabolites by Fungal Co-culture of Trichophyton rubrum and Bionectria ochroleuca. J Nat
Prod 76, 1157–1165, https://doi.org/10.1021/np400258f (2013).
27. Li, C., Wang, J., Luo, C., Ding, W. & Cox, D. G. A new cyclopeptide with antifungal activity from the co-culture broth of two marine
mangrove fungi. Nat. Prod. Res. 28, 616–621, https://doi.org/10.1080/14786419.2014.887074 (2014).
28. Xu, X. Y. et al. Metabolomics Investigation of an Association of Induced Features and Corresponding Fungus during the Co-culture
of Trametes versicolor and Ganoderma applanatum. Front Microbiol 8, 2647, https://doi.org/10.3389/fmicb.2017.02647 (2018).
29. Parrot, D., Papazian, S., Foil, D. & Tasdemir, D. Imaging the Unimaginable: Desorption Electrospray Ionization - Imaging Mass
Spectrometry (DESI-IMS) in Natural Product Research. Planta Med 84, 584–593, https://doi.org/10.1055/s-0044-100188 (2018).
30. Dunham, S. J. B., Ellis, J. F., Li, B. & Sweedler, J. V. Mass Spectrometry Imaging of Complex Microbial Communities. Acc Chem Res
50, 96–104, https://doi.org/10.1021/acs.accounts.6b00503 (2017).
31. Watrous, J., Hendricks, N., Meehan, M. & Dorrestein, P. C. Capturing Bacterial Metabolic Exchange Using Thin Film Desorption
Electrospray Ionization-Imaging Mass Spectrometry. Anal Chem 82, 1598–1600, https://doi.org/10.1021/ac9027388 (2010).
32. Gao, X. et al. Fungal indole alkaloid biosynthesis: genetic and biochemical investigation of the tryptoquialanine pathway in
Penicillium aethiopicum. J Am Chem Soc 133(8), 2729–2741, https://doi.org/10.1021/ja1101085 (2011).
33. Ariza, M. R., Larsen, T. O., Petersen, B. O., Duus, J. Ø. & Barrero, A. F. Penicillium digitatum metabolites on synthetic media and
citrus fruits. J Agric Food Chem 50, 6361–6365, https://doi.org/10.1021/jf020398d (2002).
34. Zhu, C. et al. Identification of secondary metabolite biosynthetic gene clusters associated with the infection of citrus fruit by
Penicillium digitatum. Postharvest Biol Technol, 134, 17–21, https://doi.org/0.1016/j.postharvbio.2017.07.011 (2017).
35. Raimundo, I., Silva, S. G., Costa, R. & Keller-Costa, T. Bioactive Secondary Metabolites from Octocoral-Associated Microbes – New
Chances for Blue Growth. Mar Drugs 16(12), 485, https://doi.org/10.3390/md16120485 (2018).
36. Nguyen, D. D. et al. MS/MS networking guided analysis of molecule and gene cluster families. Proc Natl Acad Sci USA 110,
E2611–E2620, https://doi.org/10.1073/pnas.1303471110 (2013).
37. Bian, Z., Marvin, C. C. & Martin, S. F. Enantioselective Total Synthesis of (-)-Citrinadin A and Revision of Its Stereochemical
Structure. J Am Chem Soc 135, 10886–10889, https://doi.org/10.1021/ja405547f (2013).
38. Bertinetti, B. V., Peña, N. I. & Cabrera, G. M. An Antifungal Tetrapeptide from the Culture of Penicillium canescens. Chem Biodivers
6, 1178–1184, https://doi.org/10.1002/cbdv.200800336 (2009).
39. Hammerl, R., Frank, O., Schmittnägel, T., Ehrmann, M. A. & Hofmann, T. Functional Metabolome Analysis of Penicillium roqueforti
by Means of Differential Off-Line LC-NMR. J Agric Food Chem 67(18), 5135–5146, https://doi.org/10.1021/acs.jafc.9b00388 (2019).
40. Sharma, K., Aaghaz, S., Shenmar, K. & Jain, R. Short Antimicrobials Peptides. Recent Pat Antiinfect Drug Discov, 13(1), 12–52,
https://doi.org/0.2174/1574891X13666180628105928 (2018).
41. Lin, Z. et al. Chrysogenamide A from an Endophytic Fungus Associated with Cistanche deserticola and Its Neuroprotective Effect on
SH-SY5Y Cells. J Antibiot 61(2), 81–85, https://doi.org/10.1038/ja.2008.114 (2008).
42. Andrade, J. A. S. et al. Citrinadin A derivates from Penicillium citrinum, an endophyte from the marine alga Dichotomaria marginata.
Planta Med 80(10), 776, https://doi.org/10.1055/s-0034-1382419 (2014).
43. Kyeremeh, K., Owusu, K. B., Ofosuhene, M., Ohashi, M. & Agyapong, J. Anti-Proliferative and Anti-Plasmodia Activity of
Quinolactacin A2, Citrinadin A and Butrecitrinadin co-isolated from a Ghanaian Mangrove Endophytic Fungus Cladosporium
oxysporum strain BRS2A-AR2F. J Chem Applications 3(1), 12 (2017).
44. Tsuda, M. et al. Citrinadin A, a Novel Pentacyclic Alkaloid from Marine-Derived Fungus Penicillium citrinum. Org Lett 6(18),
3087–3089, https://doi.org/10.1021/ol048900y (2004).
45. Hawkins, P. J., Geddes, B. A. & Oresnik, I. J. Common dyes used to determine bacterial polysaccharides on agar are affected by
medium acidification. Can J Microbiol 63(6), 559–562, https://doi.org/10.1139/cjm-2016-0743 (2017).
46. Gaulin, E. et al. The CBEL glycoprotein of Phytophthora parasitica var-nicotianae is involved in cell wall deposition and adhesion to
cellulosic substrates. J Cell Sci 115, 4565–4575, https://doi.org/10.1242/jcs.00138 (2002).
47. Mitic, M. et al. Disruption of calcineurin catalytic subunit (cnaA) in Epichloë festucae induces symbiotic defects and intrahyphal
hyphae formation. Mol Plant Pathol 16(6), 1414–1426, https://doi.org/10.1111/mpp.12624 (2017).
48. Qiu, Z., Wu, X., Gao, W., Zhang, J. & Huang, C. High temperature induced disruption of the cell wall integrity and structure in
Pleurotus ostreatus mycelia. Appl Microbiol Biotechnol 102(15), 6627–6636, https://doi.org/10.1007/s00253-018-9090-6 (2018).
49. Mazu, T. K., Bricker, B. A., Flores-Rozas, H. & Ablordeppey, S. Y. The Mechanistic Targets of Antifungal Agents: An Overview. Mini
Rev Med Chem 16(7), 555–578 (2016).
50. Georgopapadakou, N. H. & Tkacz, J. S. The Fungal cell wall as a drug target. Trends Microbiol 3(3), 98–104, https://doi.org/10.1016/
S0966-842X(00)88890-3 (1995).

Acknowledgements
This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior -
Brasil (CAPES) - Finance Code 001, Fundação de Amparo a Pesquisa no Estado de São Paulo [grant number
2017/24462-4 and 2019/06359-7], Deutsche Forschungsgemeinschaft (SFB 1127 ChemBioSys) and L’Oréal Brazil,
together with ABC and UNESCO in Brazil; CFFA was recipient of a postdoctoral Fellowship from CNPq (grant
number 400577/2015-1). TPF was recipient of a postdoctoral Fellowship from Capes/Humboldt. We would like to
thank Dr. Katia Kupper for the P. citrinum strain, Dr. Marcos Nogueira Eberlin for the orbitrap mass spectrometer
and Guerline François.

SCIENTIFIC REPORTS | (2019) 9:18647 | https://doi.org/10.1038/s41598-019-55204-9 10

www.nature.com/scientificreports/ www.nature.com/scie
124

Author contributions
J.H.C. and C.I.W. wrote the manuscript. C.F.F.A and C.I.W. performed the MSI analysis. J.H.C., T.P.F, K.S. and
C.I.W analyzed MSI data, isolated and characterized the secondary metabolites, performed the antimicrobial
assays and conducted microscopy analysis. T.P.F. oversaw the project. T.P.F., C.F.F.A., K.S. and C.H. reviewed the
manuscript. T.P.F. and J.H.C. submitted the manuscript.

Competing interests
The authors declare no competing interests.

Additional information
Supplementary information is available for this paper at https://doi.org/10.1038/s41598-019-55204-9.
Correspondence and requests for materials should be addressed to T.P.F.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre-
ative Commons license, and indicate if changes were made. The images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons license and your intended use is not per-
mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the
copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

© The Author(s) 2019

SCIENTIFIC REPORTS | (2019) 9:18647 | https://doi.org/10.1038/s41598-019-55204-9 11

 125

6.3 Informações suplementares do capítulo IV

SCIENTIFIC REPORTS

Antifungal potential of secondary metabolites involved in the interaction

between citrus pathogens

Jonas Henrique Costa1, Cristiane Izumi Wassano1, Célio Fernando Figueiredo

Angolini2, Kirstin Scherlach3, Christian Hertweck3,4, Taícia Pacheco Fill1*

1 Institute of Chemistry, University of Campinas, CP 6154, 13083-970, Campinas – SP,

Brazil
2 Center for Natural and Human Sciences, Federal University of ABC, 09210-580,
Santo André – SP, Brazil
3 Department of Biomolecular Chemistry, Leibniz Institute for Natural Product Research
and Infection Biology – Hans Knöll Institute, Jena, Germany
4 Chair of Natural Product Chemistry, Friedrich Schiller University Jena, 07743 Jena,
Germany.
*Corresponding author: phone +55-19-35213092, e-mail taicia@unicamp.br
 126

Supplementary Information
S1. Mass Spectrometry Imaging (MSI)

Figure S1. DESI-MS (+) signals obtained for tryptoquialanine A (1) on the surface of P. digitatum and
P. citrinum co-culture.

Figure S2. DESI-MS (+) signals obtained for tryptoquialanine C (2) on the surface of P. digitatum and
P. citrinum co-culture.
 127

Figure S3. DESI-MS (+) signals obtained for tryptoquialanone (3) on the surface of P. digitatum and P.
citrinum co-culture.

Figure S4. DESI-MS (+) signals obtained for 15-dimethyl-2-epi-fumiquinazoline A (4) on the surface of
P. digitatum and P. citrinum co-culture.
 128

Figure S5. DESI-MS (+) signals obtained for deoxytryptoquialanone (5) on the surface of P. digitatum
and P. citrinum co-culture.

Figure S6. DESI-MS (+) signals obtained for citrinadin A (6) on the surface of P. digitatum and P.
citrinum co-culture.
 129

Figure S7. DESI-MS (+) signals obtained for deoxycitrinadin A (7) on the surface of P. digitatum and P.
citrinum co-culture.

Figure S8. DESI-MS (+) signals obtained for Phe-Val-Val-Tyr (8) on the surface of P. digitatum and P.
citrinum co-culture.
 130

Figure S9. DESI-MS (+) signals obtained for Phe-Val-Val-Phe (9) on the surface of P. digitatum and P.
citrinum co-culture.

Figure S10. DESI-MS (+) signals obtained for chrysogenamide A (10) on the surface of P. digitatum
and P. citrinum co-culture.
 131

S2. Mass Spectrometry of in vivo co-culture

Figure S11. Extracted ion chromatograms of [M+H]+ m/z 519.18, tryptoquialanine A (1), for in vivo
extracts of (A) co-culture, (B) P. digitatum, (C) P. citrinum and in (D) orange control. (E) Mass spectrum
of 1.
 132

Figure S12. Extracted ion chromatograms of [M+H]+ m/z 503.19, tryptoquialanine C (2), for in vivo
extracts of (A) co-culture, (B) P. digitatum, (C) P. citrinum and in (D) orange control. (E) Mass spectrum
of 2.
 133

Figure S13. Extracted ion chromatograms of [M+H] + m/z 475.16, tryptoquialanone (3), for in vivo
extracts of (A) co-culture, (B) P. digitatum, (C) P. citrinum and in (D) orange control. (E) Mass spectrum
of 3.
 134

Figure S14. Extracted ion chromatograms of [M+H]+ m/z 460.19, 15-dimethyl-2-epi-fumiquinazoline A

(4), for in vivo extracts of (A) co-culture, (B) P. digitatum, (C) P. citrinum and in (D) orange control. (E)
Mass spectrum of 4.
 135

Figure S15. Extracted ion chromatograms of [M+H]+ m/z 459.16, deoxytryptoquialanone (5), for in vivo
extracts of (A) co-culture, (B) P. digitatum, (C) P. citrinum and in (D) orange control. (E) Mass spectrum
of 5.
 136

Figure S16. Extracted ion chromatograms of [M+H]+ m/z 625, citrinadin A (6), for in vivo extracts of (A)
co-culture and (B) P. citrinum and in (C) orange control. (D) Mass spectrum of 6.

Figure S17. Extracted ion chromatograms of [M+H]+ m/z 609, deoxycitrinadin A (7), for in vivo extracts
of (A) co-culture and (B) P. citrinum and in (C) orange control. (D) Mass spectrum of 7.
 137

Figure S18. Extracted ion chromatograms of [M+H]+ m/z 527, Phe-Val-Val-Tyr (8), for in vivo extracts
of (A) co-culture and (B) P. citrinum and in (C) orange control. (D) Mass spectrum of 8.

Figure S19. Extracted ion chromatograms of [M+H]+ m/z 511, Phe-Val-Val-Phe (9), for in vivo extracts
of (A) co-culture and (B) P. citrinum and in (C) orange control. (D) Mass spectrum of 9.
 138

Figure S20. Extracted ion chromatograms of [M+H]+ m/z 448, chrysogenamide A (10), for in vivo
extracts of (A) co-culture and (B) P. citrinum and in (C) orange control. (D) Mass spectrum of 10.
 139

S3. Characterization of secondary metabolites

Figure S21. MS/MS spectrum of citrinadin A (6) (30 eV).

Figure S22. MS/MS match between GNPS database (green) and ion at m/z 625 isolated from co-culture
extract (black). The tandem mass spectrum shared 5 mass fragments in common (m/z 607.38, 594.35,
576.34, 449.24 and 431.23), suggesting that ion at m/z 625 could be citrinadin A.
 140

Figure S23. MS/MS molecular networking analysis of compounds isolated from P. citrinum and P.
digitatum co-culture extract. The blue node represents a match between m/z 625 and citrinadin A in
GNPS database. The grey (m/z 609) and blue node are related metabolites grouped in same cluster
with a cosine of 0.67.

Figure S24. MS/MS spectrum of deoxycitrinadin A (7) (40 eV) and proposed fragmentation structures.
 141

Figure S25. 1H NMR spectrum of Deoxycitrinadin A (7) (600.17 MHz, DMSO).

Figure S25. COSY NMR spectrum of Deoxycitrinadin A (7) (600.17 MHz, DMSO).
 142

Table S1. 1H NMR data and 1H-1H correlations in COSY for 7 (600 MHz, DMSO) and comparison with
the literature (δ in ppm, J in Hz).

Position 1
H δ (m, J) H δ (m, J)1
1
COSY
1 8.52 (s) 9.77 (s)
4 7.54 (d, 7.0) 7.57 (d, 7.2) H-5
5 7.15 (t, 7.0) 7.06 (dd, 7.8, 7.2) H-4, H-6
6 7.79 (d, 8.1) 7.67 (dd, 7.8, 0.6) H-5
8 1.98, 1.94 (ABq, 11.5) 2.08, 2.04 (ABq, 14.4)
10 3.00 (d, 10.9) 3.15 (d, 10.8)
10 2.53 (d, 10.9) 2.54 (d, 10.8)
12 3.03 – 2.99 (m) 3.01 (quin, 6.6) H-27, H-13
13 1.58 – 1.55 (m) 1.50 – 1.45 (m) H-12, H-14
13 2.03 – 1.99 (m) 2.04 – 1.98 (comp)
14 5.10 – 5.08 (m) 5.19 – 5.20 (m) H-13, H-15
15 1.70 – 1.66 (m) 1.80 – 1.76 (comp) H-14, H-16
15 1.81 – 1.79 (m) 1.80 – 1.76 (comp)
16 3.18 – 3.17 (m) 3.20 – 3.14 (m) H-15, H-17
17 1.39 – 1.38 (m) 1.36 – 1.31 (comp) H-16
17 1.58 – 1.55 (m) 1.53 (dd, 13.2, 4.2)
18-OH 4.66 (s) 4.70 (d, 2.4)
21 6.91 – 6.90 (m) 6.75 – 6.74 (m) H-24, H-23
23 2.15 (s) 2.17 (d, 1.2) H-21
24 2.01 (s) 2.01 (d, 1.2) H-21
26 2.24 – 2.23 (m) 2.28 (s)
27 1.15 (d, 7.1) 1.19 (d, 7.2) H-12
28 0.85 (s) 0.96 (s)
29 1.24 (s) 1.36 – 1.31 (comp)
2’ 2.65 (d, 10.9) 2.68 (d,10.8) H-4’
4’ 1.95 – 1.93 (m) 2.04 – 1.98 (comp) H-2’, H-5’, H-6’
5’ 0.92 (d, 6.9) 0.95 (d, 6.6) 4’
6’ 0.85 (d, 6.4) 0.88 (d, 6.6) 4’
7’ 2.23 (s) 2.29 (s)
8’ 2.23 (s) 2.29 (s)
1 Bian, Z., Marvin, C. C., Martin, S. F. Enantioselective Total Synthesis of (-)-Citrinadin A and Revision
of Its Stereochemical Structure. J Am Chem Soc 135, 10886-10889, https://doi.org/10.1021/ja405547f
(2013).
 143

Figure S27. MS/MS spectrum of Tyr-Val-Val-Phe (8).

Figure S28. 1H NMR spectrum of Tyr-Val-Val-Phe (8) (500.13 MHz, DMSO).

Figure S29. 13C NMR spectrum of Tyr-Val-Val-Phe (8) (500.13 MHz, DMSO).
 144

Table S2. 1H and 13C NMR data for 8 (500 MHz, DMSO) (δ in ppm, J in Hz).

Position 1H δ (m, J) 13C δ

Phe
1 - 160.0
2 4.24 (t, J = 7.6) 53.2
3 3.04 (dd, J = 14.0, 7.0) 37.5
2.89 – 2.95 (m)
4 - 134.9
5,9 7.23 – 7.33 (m) 129.5
6, 8 7.23 – 7.33 (m) 127.4
7 7.23 – 7.33 (m) 127.2
Val
1’ - 170.7
2’ 4.32 (dd, J = 9.4, 5.7) 56.8
3’ 1.80 – 1.86 (m) 31.6
4’ 0.67 (d, J = 7.0) 19.0
5’ 0.52 (d, J = 6.8) 17.2
NH 8.32 (d, J = 8.3) -
Val
1” - 170.1
2” 4.54 (dd, J = 9.4, 5.0) 56.7
3” 1.73 – 1.79 (m) 31.0
4” 0.66 (d, J = 6.8) 19.2
5” 0.63 (d, J = 6.9) 17.2
NH 8.45 (d, J = 9.1) -
Tyr
1”’ - 173.3
2”’ 4.34 – 4.39 (m) 53.4
3”’ 2.89 – 2.95 (m) 36.2
2.66 (dd, J = 14.0, 9.7)
4”’ - 128.5
5’’’,9”’ 6.99 (d, J = 8.5) 130.0
6’’’, 8”’ 6.61 (d, J = 9.1) 114.9
7”’ - 155.9
NH 8.10 (d, J = 9.2) -
 145

Figure S30. MS/MS spectrum of Phe-Val-Val-Phe (9).

Figure S31. 1H NMR spectrum of Phe-Val-Val-Phe (9) (500.13 MHz, DMSO).

Figure S32. 13C NMR spectrum of Phe-Val-Val-Phe (9) (500.13 MHz, DMSO).
 146

Table S3. 1H and 13C NMR data for 9 (500 MHz, DMSO) (δ in ppm, J in Hz).

Position 1
H δ (m, J) 13
Cδ
Phe
1 - 168.1
2 4.21 – 4.25 (m) 53.3
3 3.04 (dd, J = 6.1, 13.7) 37.0
2.77 ( J = 10.7, 13.8)
4 - 134.9
5,9 7.21 – 7.33 (m) 129.2
6, 8 7.21 – 7.33 (m) 128.6
7 7.21 – 7.33 (m) 127.3
Val
1’ - 170.2
2’ 4.31 (dd, J = 5.5, 9.2) 56.7
3’ 1.71 (m) 31.0
4’ 0.63 (d, J = 6.8) 19.1
5’ 0.60 (d, J = 6.8) 19.2
NH 8.45 (d, J = 10.2) -
Val
1” - 170.7
2” 4.54 (dd, J = 4.8, 9.2) 56.9
3” 1.82 (m) 31.6
4” 0.66 (d, J = 7.2) 17.3
5” 0.45 (d, J = 7.0) 17.3
NH 8.39 (d, J = 8.6) -
Phe
1”’ - 173.2
2”’ 4.47 (ddd, J = 4.3, 8.5, 18.7) 53.4
3”’ 3.07 (dd, J = 4.4, 14.1) 37.6
2.91 (dd, J = 8.5, 14.0)
4”’ - 137.5
5’’’,9”’ 7.21 – 7.33 (m) 129.5
6’’’, 8”’ 7.21 – 7.33 (m) 128.2
7”’ 7.21 – 7.33 (m) 126.5
NH 8.01 (s) -
 147

Figure S33. 1H NMR spectrum of chrysogenamide A (10) (500.13 MHz, DMSO).

Figure S34. 13C NMR spectrum of chrysogenamide A (10) (500.13 MHz, CDCl3).
 148

Table S4. 1H and 13C NMR data for 10 (500 MHz, DMSO) (δ in ppm, J in Hz).

10

Position 1
H δ (m, J) 13
Cδ
1-NH 10.52 (s) -
2 - 183.1
3 - 72.6
4 7.19 (d, J = 7.2) 126.4
5 6.94 (d, J = 7.7) 121.9
6 6.98 (d, J = 7.4) 128.3
7 - 123.6
8 - 140.4
9 - 129.7
10 40.0
11 - 69.9
12a 3.30 (d = 5.8) 63.1
12b 2.39 (m) -
13 - 65.8
14a 1.49 (m) 26.6
14b 1.15 (m) -
15a 2.09 (m) 29.2
15b 10.52 (s) -
16a 1.57 (m) 35.2
16b 1.08 (m) -
17 1.99 (m) 65.8
18 - 176.5
19-NH 9.23 (s) -
20a 1.62 (m) 36.3
20b 1.28 (m) -
21 2.62 (m) 46.7
22 - 46.7
23 1.23 (d, J = 6.9) 22.1
24 0.66 (s) 20.5
25 0.91 (s) 20.5
26 3.24 (m) 28.4
27 5.27 (tt, J = 7.3, 1.3) 122.7
28 - 132.6
29 1.70 (s) 25.5
30 1.67 (s) 17.8
 149

S4. Antifungal assays

Figure S35. P. digitatum growing on (A) PDA and (B) PDA with 0.5 mg ml-1 of the co-cultive extract. P.
digitatum exhibited reduction in radial growth of 67% in presence of P. citrinum metabolites

Figure S36. Microplate of MIC assay of tryptoquialanine A (1) and 15-dimethyl-2-epi-fumiquinazoline A

(4) against P. citrinum. *NC is the negative control (no inoculation). **FG is the fungal control in YES
media and 5% methanol. ***PC is the positive control (50 µg mL-1 of itraconazole).
 150

S5. Confocal Laser Scanning Microscopy

Figure S37. Confocal laser scanning microscopy of Congo Red-stained co-culture of P. digitatum and
P. citrinum. (A) Interface zone between PD (below) and PC (above). PD hyphae: (B) sample and (C)
control. Zoom in: (D) PD sample and (E) PD control. (F) Comparison between PD hyphae in control
(above) and in sample (below): patches of Congo Red indicates a defective fungal cell wall. Bars = 5.0
µm
 151

7 DISCUSSÃO
Apesar da importância de P. digitatum para a citricultura, muitos aspectos dos
mecanismos de infecção ainda permanecem desconhecidos 21. Ainda, a maioria dos
trabalhos sobre P. digitatum focam no tratamento dos sintomas dos frutos doentes,
resistência a fungicidas e no biocontrole do bolor verde através do uso de
microrganismos antagonistas21. Uma vez que o entender a relação patógeno-
hospedeiro é fundamental para o desenvolvimento de estratégias mais seguras para
o controle de doenças pós-colheita, os trabalhos apresentados nesta tese focam no
estudo do processo de infecção do fungo. Os trabalhos publicados visam
compreender o papel biológico de metabólitos secundários, as respostas de defesa
dos hospedeiros e a especificidade sobre o hospedeiro citros.
As técnicas de IMS e redes moleculares foram adequadamente utilizadas,
de forma que permitiram obter dados importantes que auxiliaram no entendimento da
relação patógeno-hospedeiro. IMS se tornou uma ferramenta importante para a área
de produtos naturais, principalmente para microbiologistas, uma vez que fornece a
distribuição espacial de diversas moléculas22. Basicamente a técnica é baseada na
dessorção e ionização de analitos de uma superfície, seguido da análise de massas
dos íons da fase gasosa resultante. Esse processo é repetido várias vezes, de forma
a se associar os espectros de massa com cada posição na amostra. Os íons são
correlacionados com sua distribuição e a abundância molecular é indicada com a
intensidade das cores, permitindo a criação de uma imagem 2D23,24.
A técnica de IMS pode ser dividida em três etapas. A primeira etapa
consiste na preparação e ionização da amostra a ser analisada e várias são as
técnicas de ionização que foram desenvolvidas para esse propósito24. Secondary-ion
mass spectrometry (SIMS), Matrix assisted laser-desorption ionization (MALDI) e
Desorption electrospray ionization (DESI) são algumas das técnicas aplicadas com
sucesso em imageamento, sendo também as mais comuns e utilizadas 24.
A técnica de DESI foi a utilizada nas análises de IMS apresentadas nesta
tese. DESI é feita em condições ambientes e possui vantagens que a torna adequada
para a utilização em produtos naturais23. Algumas dessas vantagens são a ionização
externa ao espectrômetro de massas, a mínima ou nula preparação da amostra, a
mínima destruição da amostra (o que permite a repetição das análises) e o fato de a
amostra não ser introduzida dentro do espectrômetro de massas 23-26. Essas
propriedades tornam DESI-IMS uma ferramenta poderosa no estudo das interações
 152

entre microrganismos ou patógeno-hospedeiro23. Na literatura, outras interações entre

fitopatógenos-hospedeiro também foram analisadas por DESI-IMS com o intuito
desvendar a resposta de plantas em relação ao ataque de patógenos 23. Tata et al.
(2015) verificou que o metabolismo de batatas infectadas com o fungo Pythium
ultimum foi alterado, observando aumento e diminuição do nível de 12 glicoalcalóides
e a produção de algumas toxinas27.
Nos capítulos I, III e IV percebe-se que DESI-IMS foi fundamental no
entendimento do papel biológico dos metabólitos secundários produzidos por P.
digitatum e no descobrimento de novos compostos. No capítulo I, DESI-IMS foi capaz
de detectar na superfície de laranjas infectadas com P. digitatum metabólitos da rota
biossintética das triptoquialaninas como triptoquialanina A, triptoquialanina B,
triptoquialanona, 15-dimetil-2-epi-fumiquinazolina A e deoxitriptoquialanona. Além
desses metabólitos, triptoquivalina L e deoxitriptoquialanina foram detectadas pela
primeira vez para P. digitatum. As análises de IMS também mostraram um acúmulo
dos alcaloides na superfície dos frutos de 4 até 7 dias após a infecção. Este resultado
é compatível com análises de RNA-seq reportadas na literatura que indicam um
aumento da expressão dos genes biossintéticos das triptoquialaninas com o
desenvolvimento da infecção28. Através do acúmulo das triptoquialaninas na
superfície das frutas detectado por IMS, verificou-se a possibilidade destes
metabolitos estarem envolvidos na proteção do sistema patógeno-hospedeiro contra
a invasão de insetos. Esta hipótese foi confirmada através de ensaios que
confirmaram a ação inseticida da triptoquialanina A.
No capítulo IV, IMS também foi fundamental para desvendar o papel
biológico dos metabólitos fúngicos. As triptoquialaninas, tetrapeptídeos,
chrisogenamidas e citrinadinas, detectadas na zona de confronto entre P. digitatum e
P. citrinum, tiveram suas ações antifúngicas confirmadas em ensaios antimicrobianos.
Os compostos produzidos por P. citrinum inibiram o crescimento radial de P. digitatum
e ainda causaram danos na parede celular fúngica. A parede celular é um alvo atrativo
quando se trata de antifúngicos, uma vez que não está presente nas células animais 29,
o que torna estes compostos potenciais agentes antifúngicos que merecem futura
investigação.
Vale ressaltar que até os estudos realizados nesta tese, não havia na
literatura nenhuma atividade biológica reportada para as triptoquialaninas, metabólitos
que são majoritariamente produzidos por P. digitatum. Essas atividades biológicas só
 153

foram possíveis com a utilização das informações espaciais obtidas por DESI-IMS
tanto no sistema fungo-hospedeiro quanto no sistema fungo-fungo.
Visto que DESI-IMS é uma técnica limitada a superfície da amostra, os
metabólitos internos não são detectados22,30, e são necessárias técnicas
complementares para a identificação dos metabólitos secundários de interesse como,
por exemplo, o uso de bancos de dados online de produtos naturais e redes
moleculares (Molecular Networking).
A desreplicação de extratos naturais geralmente envolve uma avaliação
das massas exatas e do padrão de fragmentação dos compostos obtidos por
espectrometria de massas31. Entretanto, uma simples análise de massas de um único
extrato pode gerar milhões de espectros, e uma análise manual destes é inviável
quando pensamos no tempo que será consumido para o processamento dos
dados30,31,32.
A fragmentação de uma molécula gera um espectro tandem (MS/MS) que
é único para essa estrutura, como uma impressão digital30. Desta forma, os padrões
de fragmentação obtidos na metabolômica podem ser comparados com espectros
MS/MS de compostos conhecidos em bancos de dados online como, por exemplo,
METLIN, MassBank e GNPS (Global Natural Products Social Molecular
Networking)30,31-34, facilitando o processamento dos dados obtidos. O GNPS, por
exemplo, é uma plataforma que permite a desreplicação online através de espectros
MS/MS depositados pela comunidade32.
A plataforma GNPS também pode ser utilizada para se obter o molecular
networking32, uma ferramenta empregada para a rápida desreplicação de extratos de
produtos naturais35. O molecular networking utiliza espectros de MS/MS em alta
resolução, organizando-os em uma representação visual na forma de uma rede
molecular31,35. Um dos benefícios do molecular networking é que, uma vez que os
compostos são agrupados levando em consideração a fragmentação, a identificação
de compostos semelhantes se torna mais fácil30. Além disso, a visualização da rede
permite a comparação do perfil metabólico de diferentes extratos em diferentes
tempos, condições ou interações30,35. Assim, a utilização das redes moleculares, com
o auxílio de identificação de metabólitos na plataforma GNPS, permitiu a rápida
desreplicação dos dados obtidos para P. digitatum, facilitando também a comparação
do perfil metabólito fúngico em diferentes hospedeiros.
 154

No capítulo I e II, as redes moleculares mostram que a produção das

triptoquialaninas e seus intermediários não é alterada, de forma que o fungo produz
estes alcaloides em limão siciliano, laranja, tangerina, suco de laranja, ameixa e meio
de cultura sintético. Estes resultados reforçam a importância das triptoquialaninas
para P. digitatum durante o processo de infecção. No capítulo III, as comparações
entre as redes moleculares obtidas para as VEs isoladas de P. digitatum e para as
infecções nos frutos permitiu visualizar que apenas os metabólitos finais da rota
biossintética das triptoquialaninas estavam presentes nas VEs. Esta constatação, em
conjunto com a quantificação das triptoquialaninas, foi crucial para sugerir que as
triptoquialaninas são biossintetizadas nas células fúngicas e apenas transportadas
pelas VEs. No capítulo IV, através da comparação das redes moleculares entre frutos
saudáveis e infectados foi possível confirmar os flavonoides hidroxilados pela planta
observados por DESI-IMS. Ainda, o banco de dados do GNPS auxiliou na
identificação de compostos nunca reportados para P. digitatum, como triptoquivalinas
L e Q, e fumiquinazolinas A e C.
Além das atividades inseticidas e antimicrobianas observadas para as
triptoquialaninas, no capítulo II observou-se que as triptoquialaninas possuem
atividade fitotóxica sob sementes de Citrus sinensis. Através de ensaios de
germinação observou-se que 100% das sementes expostas a uma concentração de
3000 ppm de triptoquialanina A não germinaram. Os resultados foram confirmados
também através de gráficos de PCA e PLS-DA que mostraram uma diferença
metabólica entre as sementes expostas ao alcaloide e as sementes controles tratadas
com água. Desta forma, os resultados apresentados nesta tese mostraram que as
triptoquialaninas são um metabólito importante para P. digitatum durante a infecção.
Um estudo anterior reportou, através da deleção do gene biossintético das
triptoquialaninas, que mutantes deficientes na produção de triptoquialaninas ainda
eram capazes de infectar os frutos, de forma que estes metabólitos não são essenciais
para a virulência de P. digitatum28. Assim, acreditava-se que as triptoquialaninas não
eram importantes durante a infecção. Esses alcaloides aparentam estar envolvidos
nos mecanismos de ataque e defesa do fungo, defendendo o sistema fungo-
hospedeiro de invasores como insetos e outros microrganismos competidores e
atacando o sistema hospedeiro através de ações fitotóxicas.
Todas as atividades biológicas observadas para as triptoquialaninas
também levaram a investigação do mecanismo de transporte destes alcaloides. Uma
 155

vez que as atividades antimicrobianas, fitotóxicas e inseticidas ocorrem no meio

extracelular, levantou-se a hipótese de que as VEs estariam envolvidas no transporte
das triptoquialaninas de dentro das células fúngicas até o meio extracelular. Pela
primeira vez, VEs foram isoladas para P. digitatum tanto no sistema in vitro quanto in
vivo. Em vários sistemas patógeno-hospedeiro, as VEs são consideradas fatores
determinantes na patogenicidade uma vez que carregam fatores virulência 36,37. Até o
momento, resultados semelhantes só haviam sido obtidos para o fitopatógeno
Fusarium oxysporum f. sp. vasinfectum, cujas vesículas causaram lesões em folhas
de algodão38. Esses resultados abrem novas linhas de investigação em relação a
patogenicidade de P. digitatum.

8 CONCLUSÕES E PERSPECTIVAS
Conclui-se que esta tese de doutorado cumpriu com o principal objetivo
proposto: avançar no entendimento dos mecanismos de patogenicidade do fungo P.
digitatum através do estudo dos metabólitos secundários envolvidos na interação
patógeno-hospedeiro. Este objetivo foi alcançado através do uso de metodologias
poderosas na pesquisa em produtos naturais, como as técnicas de imageamento por
espectrometria de massas, redes moleculares e co-culturas. As técnicas
moencionadas foram empregadas com sucesso nos quatro artigos científicos que
fazem parte desta tese.
Nos capítulos I, II e IV, estudos de bioatividade mostraram que a
triptoquialanina A, metabólito secundário produzido majoritariamente por P. digitatum,
é importante durante o processo de infecção, possuindo atividade inseticida,
antimicrobiana e fitotóxica. Apesar de não ser indispensável para a patogenicidade, a
triptoquialanina A aparenta estar envolvida na proteção P. digitatum contra insetos e
outros microrganismos que competem pelo hospedeiro citros. Vale ressaltar que as
triptoquialaninas não tinham nenhuma função biológica descrita na literatura até a
publicação do estudos apresentados nesta tese.
No capítulo II, a investigação sobre o transporte das triptoquialaninas levou
à descoberta de vesículas extracelulares (VEs) fitotóxicas produzidas por P. digitatum.
Observou-se que as VEs de P. digitatum transportavam triptoquialaninas e também
outras micotoxinas. A associação entre VEs e metabólitos secundários nunca havia
sido descrita na literatura até o momento. Visto que em vários sistemas patógeno-
hospedeiro as VEs carregam fatores de virulência, a descoberta das vesículas de P.
 156

digitatum abre um novo capítulo no entendimento das estratégias de virulência deste

fitopatógeno. A carga enzimática, função e biogênesis das VEs de P. digitatum são
exemplos de tópicos que ainda precisam ser investigados e, portanto, oferecem
oportunidade de estudo.
No capítulo III, os mecanismos de patogenicidade de P. digitatum foram
avaliados através da interação enzimática entre fitopatógeno e flavonóides cítricos. A
interação entre fitopatógenos e flavonóides já era conhecida na literatura, mas nunca
havia sido completamente explorada. Comparando-se diversos fungos fitopatógenos,
P. digitatum e P. italicum, os patógenos mais agressivos frente à citros, mostraram
maior taxa de hidrólise dos flavonóides naringina e hesperidina. Ainda, foram
detectados novos compostos envolvidos na defesa das plantas contra os
fitopatógenos. O artigo apresentado neste capítulo fornece mais evidências sobre a
resposta fúngica contra os mecanismos de defesa das plantas, abrindo novas
oportunidades de investigação.
Por fim, o capítulo IV mostrou que a técnica de co-cultivo envolvendo
fitopatógenos é uma ótima estratégia na busca de meios mais seguros para o controle
da doença causa pelo bolor verde. Os metabólitos secundários produzidos pelo fungo
antagonista à P. digitatum foram capazes de inibir o crescimento do bolor verde in vivo
e in vitro. A utilização destas moléculas para o biocontrole da doença do bolor verde
necessita mais estudos, uma vez que se mostra uma excelente alternativa ao uso dos
fungicidas sintéticos que, atualmente, são o principal meio de controle desta doença.
Em resumo, os trabalhos desenvolvidos durante esta tese visam contribuir
para a compreensão da interação entre P. digitatum e seus hospedeiros. Essas
contribuições podem levar ao desenvolvimento de técnicas mais seguras e eficientes
no controle da doença do bolor verde, seja através de estratégias em resposta aos
mecanismos de virulência fúngicos ou pela utilização de produtos naturais em
alternativa aos pesticidas sintéticos.
 157

9 REFERÊNCIAS BIBLIOGRÁFICAS
1 Viegas, C., Bolzani, V.S., Barreiro, E.J., 2006. Os produtos naturais e química
medicinal moderna. Quim. Nova, 29(2), 326-337.
2 Berlinck, R.G.S., Borges, W.S., Scotti, M.T., Vieira, P.C., 2017. A química de
produtos naturais do Brasil do século XXI. Quim. Nova, 40(6), 706-710.
3 Pye, C.R., Bertin, M.J., Lokey, R.S., Gerwick, W.H., Linington, R.G., 2017.
Retrospective analysis of natural products provides insights for future discovery trends.
Proc. Natl. Acad. Sci. U.S.A., 114(22), 5601-5606.
4 Li, G., Lou, H., 2017. Strategies to diversify natural products for drug discovery. Med
Res Rev, 38, 1255- 1294.
5 Demain, A.L., 2014. Importance of microbial natural products and the need to
revitalize their discovery. J. Ind. Microbiol Biotechnol, 41,185-201
6 Khan, R.A., 2018. Natural products chemistry: The emerging trends and prospective
goals. Saudi Pharm. J., 26,739–753.
7 Bishayee, A., Sethi, G., 2016. Bioactive natural products in cancer prevention and
therapy: Progress and promise. Semin. Cancer Biol., 40-41, 1-3.
8 De Oliveira, L.G., Pupo, M.T., Vieira, P.C., 2013. Explorando produtos naturais
microbianos na fronteira da química e da biologia. Quim. Nova, 36(10), 1577-1586.
9 Ghooshkhaneh, G., N., Golzarian, M. R., Mamarabadi, M., 2018. Detection and
classification of citrus green mold caused by Penicillium digitatum using multispectral
imaging. Journal of the Science of Food and Agriculture. 98, 3542-3550.
10 Jr-Mattos D, Carlos EF. 2017. The role of the International Society of Citriculture on
the world citrus industry. Citrus R&T 38(2), 228-232.
11 Citrus: World Markets and Trade. United States Department of Agriculture. Foreign
Agricultural Service, January, p. 1–13, 2020. Available online:
https://downloads.usda.library.cornell.edu/usda-
esmis/files/w66343603/00000g55g/kp78h0193/citrus.pdf (acessado em 30 de
novembro de 2020).
12 Lima FB, Félix C, Osório N, Alves A, Vitorino R, Domingues P, Correia A, Ribeiro
RTS, Esteves AC. 2016. Secretome analysis of Trichoderma atroviride T17 biocontrol
of Guignardia citricarpa. Biol Control 99, 38-86.
13 Qian X, Yang Q, Zhang Q, Abdelhai MH, Dhanasekaran S, Serwah BNA, Gu N,
Zhang H. 2019. Elucidation of the Initial Growth Process and the Infection Mechanism
 158

of Penicillium digitatum on Postharvest Citrus (Citrus reticulata Blanco).

Microorganisms 7, 485.
14 Chen J, Shen Y, Chen C, Wan C. 2019. Inhibition of Key Citrus Postharvest Fungal
Strains by Plant Extracts In Vitro and In Vivo: A Review. Plants 8, 26.
15 Macarisin, D., Cohen, L., Eick, A., Rafael, G., Belausov, E., Wisniewski, M., Droby,
S., 2007. Penicillium digitatum suppresses production of hydrogen peroxide in host
tissue during infection of citrus fruit. Phytopathology. 97, 1491-1500.
16 Hao, W., Li, H., Hu, M., Yang, L., Rizwan-ul-Haq, M., 2011. Integrated control of
citrus green and blue mold and sour rot by Bacillus amyloliquefaciens in combination
with tea saponin. Postharvest Biol. Technol. 59, 316-323.
17 Kanetis, L., Förster, H., Adaskaveg, J.E., 2010. Determination of natural resistance
frequencies in Penicillium digitatum using a new air-sampling method and
characterization of Fludioxonil- and Pyrimethanil-Resistant isolates. Phytopathol. 100,
738-743.
18 Newman, D. J., Cragg, G. M., 2012. Natural Products As Sources of New Drugs over
the 30 Years from 1981 to 2010". J. Nat. Prod, 75, 311.
19 Morrissey, J.P., Osbourn, A. E., 1999. Fungal resistance to plant antibiotics as a
mechanism of pathogenesis. Microbiology and Molecular Biology Reviews, 63(3), 708-
724.
20 Marcet-Houben, M., Ballester, A.R., la Fuente, B., Harries, E., Marcos, J.F.,
González-Candelas, L., Galbadón, T., 2012. Genome sequence of the necrotrophic
fungus Penicillium digitatum, the main postharvest pathogen of citrus. BMC Genomics.
13:646.
21 Costa J.H., Bazioli J.M., Pontes, J.G., Fill, T.P. 2019 Penicillium digitatum infection
mechanisms in citrus: What do we know so far? Fungal Biol. 123 (8), 584–593.
22 Sica, V.P., Raja, H.A., El-Elimat, T., Oberlies, N.H., 2014. Mass spectrometry
imaging of secondary metabolites directly on fungal cultures. RSC Adv., 4, 63221-
63227.
23 Parrot, D., Papazian, S., Foil, D., Tasdemir, D., 2017. Imaging the Unimaginable:
Desorption Electrospray Ionization - Imaging Mass Spectrometry (DESI‑IMS) in
Natural Product Research. Planta Med., 84, 584-593.
24 Cabral, E.C., Mirabelli, M.F., Perez, C.J., Ifa, D.R., 2013. Blotting Assisted by
Heating and Solvent Extraction for DESI-MS Imaging. J. Am. Soc. Mass Spectrom.,
24, 956-965.
 159

25 Freitas, J.R.L., Vendramini, P.H., Melo, J.O.F., Eberlin, M.N., Augusti, R., 2018. An
Appraisal on the Source-to-Sink Relationship in Plants: an Application of Desorption
Electrospray Ionization Mass Spectrometry Imaging. J. Braz. Chem. Soc., 29(1), 17-
23.
26 Wu, C., Dill, A.L., Eberlin, L.S., Cooks, R.G., Ifa, D.R., 2013. Mass Spectrometry
Imaging under ambient conditions. Mass Spectrom. Rev., 32, 218-243.
27 Tata, A., Perez, C.J., Hamid, T.S., Bayfield, M.A., Ifa, D.R., 2015. Analysis of
metabolic changes in plant pathosystems by imprint imaging DESI‑MS. J. Am. Soc.
Mass Spectrom., 26, 641–648.
28 Zhu, C., Sheng, D., Wu, X., Wang, M., Hu, X., Li, H., Yu, D. 2017. Identification of
secondary metabolite biosynthetic gene clusters associated with the infection of citrus
fruit by Penicillium digitatum. Postharvest Biol Tec 134, 17-21.
29 Mazu, T. K., Bricker, B. A., Flores-Rozas, H. & Ablordeppey, S. Y. 2016. The
Mechanistic Targets of Antifungal Agents: An Overview. Mini Rev Med Chem 16(7),
555–578.
30 Watrous, J., Roach, P., Alexandrov, T., Heath, B.S., Yang, J.N., Kersten, R.D.,
Voort, M., Pogliano, K., Gross, H., et al., 2012. Mass spectral molecular networking of
living microbial colonies. Proc. Natl. Acad. Sci. U.S.A., 109(26), E1743-E1752.
31 Krug, D., Muller, R., 2014. Secondary metabolomics: the impact of mass
spectrometry-based approaches on the discovery and characterization of microbial
natural products. Nat. Prod. Rep.,31, 768-783.
32 Wang, M., Carver, J.J., Phelan, V.V., Sanchez, L.M., et al., 2016. Sharing and
community curation of mass spectrometry data with Global Natural Products Social
Molecular Networking. Nature Biotechnol., 34, 828-837.
33 Gaudêncio, S.P. and Pereira, F., 2015. Dereplication: racing to speed up the natural
products discovery process. Nat. Prod. Res. 32, 779-810.
34 Smith, C.A., O'Maille G., Want, E.J., Qin, C., Trauger, S.A., Brandon, T.R., Siuzdak,
G., 2005. METLIN: a metabolite mass spectral database. Ther. Drug Monit., 27, 747-
751.
35 Purves, K., Macintyre, L., Brennan, D., Kuttner, E., Young, L.C., Green, D.H.,
Edrada-Ebel, R., Duncan, KR., et al., 2016. Using Molecular Networking for Microbial
Secondary Metabolites Bioprospecting. Metabolites, 6(1).
 160

36 Souza JAM, Baltazar LM, Carregal VM, Gouveia-Eufrasio L, et al. 2019.

Characterization of Aspergillus fumigatus Extracellular Vesicles and Their Effects on
Macrophages and Neutrophils Functions. Front Microbiol 10:2008.
37 Herkert PF, Amatuzzi RF, Alves LR, Rodrigues ML. 2019. Extracellular Vesicles as
Vehicles for Delivery of Biologically Active Fungal Molecules. Curr Protein Pept Sci
20(10), 1027-1036.
38 Bleackley MR, Samuel M, Garcia-Ceron D, McKenna JA, Lowe, RGT, Pathan M,
Zhao K, A C, Mathivanan S, Anderson MA. 2020. Extracellular Vesicles From the 88
Cotton Pathogen Fusarium oxysporum f. sp. vasinfectum Induce a Phytotoxic
Response in Plants. Front Plant Sci 10:1610.
 161

10 ANEXOS

10.1 Licença de reprodução do artigo de revisão e dos capítulos I e III

Figura A1. Licença de reprodução da editora Elsevier para inclusão de artigos em tese.
Disponível em: https://www.elsevier.com/about/policies/copyright/permissions (Acessado em
30 de novembro de 2020).

Figura A2. Licença de reprodução da editora Elsevier para uso pessoal de artigos. Disponível
em: https://www.elsevier.com/about/policies/copyright (Acessado em 30 de novembro de
2020).
 162

10.2 Licença de reprodução do capítulo II e IV

Figura A3. Resumo da licença de reprodução dos capítulos II e IV pela CC BY 4.0. A cópia
da licença completa está disponível em: https://creativecommons.org/licenses/by/4.0/
(Acessado em 30 de novembro de 2020).

Figura A4. Licença de reprodução do capítulo II pela mBio. Disponível em:

https://mbio.asm.org/content/author-warranty-and-provisional-license-publish (Acessado em
30 de novembro de 2020).

Figura A5. Licença de reprodução do capítulo IV pela Scientific Reports. Disponível em:
https://www.nature.com/srep/journal-policies/editorial-policies#license-agreement (Acessado
em 30 de novembro de 2020).
 163

10.3 Autorização da Comissão Interna de Biossegurança (CIBio-IQ)

 164

10.4 Autorização de acesso ao patrimônio genético (SISGEN)

 165

10.5 Declaração de tese em formato alternativo

DECLARAÇÃO

As cópias dos documentos de minha autoria ou de minha coautoria, já publicados ou

submetidos para publicação em revistas científicas ou anais de congressos sujeitos a
arbitragem, que constam da minha Tese de Doutorado, intitulada “Estudo metabólico
e dos mecanismos de patogenicidade do fungo Penicillium digitatum frente ao seu
hospedeiro citros” não infringem os dispositivos da Lei nº 9.610/98, nem o direito
autoral de qualquer editora.

Campinas, 09 de fevereiro de 2021.

Jonas Henrique Costa

R.G. nº 48.559.040-2

Taicia Pacheco Fill

RG nº 44.353.907-8