Escolar Documentos
Profissional Documentos
Cultura Documentos
CAMPINAS
2021
Ficha catalográfica
Universidade Estadual de Campinas
Biblioteca do Instituto de Química
Simone Luiz Alves - CRB 8/9094
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
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
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
7 DISCUSSÃO.........................................................................................................151
8 CONCLUSÕES E PERSPECTIVAS......................................................................155
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
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).
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.
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
Fungal Biology
journal homepage: www.elsevier.com/locate/funbio
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.
https://doi.org/10.1016/j.funbio.2019.05.004
1878-6146/© 2019 British Mycological Society. Published by Elsevier Ltd. All rights reserved.
18
J.H. Costa et al. / Fungal Biology 123 (2019) 584e593
color with spore production (Vu et al., 2018; Ismail and Zhang, considerable accumulation of H2O2 in the host (Macarisin et al.,
2004). 2007). In addition, the authors correlate the effect of organic
Currently, the postharvest control is carried out by the mass acids and low pH buffer with H2O2 host production based on the
application of fungicides in commercialized fruits. A number of evaluation of whether the accumulation of certain organic acids
synthetic fungicides, such as prochloraz, imazalil, and pyrimethanil, might result in inhibition of the oxidative burst. Treatment with
have been used worldwide as postharvest treatments to control the citric acid, oxalic acid and ascorbic acid contributed the most to the
pathogen in citrus fruits (Hao et al., 2011). Nevertheless, the increase in pathogen virulence and infection severity (Macarisin
repeated use of certain fungicides has led to the appearance of et al., 2007).
fungicide-resistant populations of P. digitatum, representing a sig-
nificant obstacle to postharvest conservation (Kanetis et al., 2010). 2.2. Catalase production as a fungal infection strategy
In addition to the serious implications for product toxicity, envi-
ronmental risk and low consumer confidence (Hao et al., 2011). Macarisin et al. (2007) also evaluated the effects of catalase
Alternative strategies to control P. digitatum infection have been addition (CAT) on the infection incidence and lesion diameters.
investigated. Recently, Lafuente et al. (2018) verified the effect of These researchers reported that 500 U/mL catalase was sufficient to
light-emitting diode (LED) blue light (LBL) at a 450-nm wavelength promote a significant increase in the lesion diameter of samples
reducing the viability and the capacity of the infection by infected with P. digitatum and, to a lesser extent, with P. expansum
P. digitatum in citrus fruits. The application of noncontinuous LBL is (Macarisin et al., 2007). Catalase is an antioxidant enzyme pro-
a promising alternative to decrease the capacity of postharvest duced by fungi (Chaga et al., 1992; Aguirre et al., 2006; Ballester
decay since it is responsible for anomalous fungal spore morphol- et al., 2006), and it decomposes hydrogen peroxide to water and
ogies and darkening of the cytoplasm. In addition, a significant oxygen. Therefore, it is suggested that the H2O2 that is produced by
decrease in ethylene production in plates that were exposed to the the plant as a defense mechanism is decomposed by the catalase
continuous LBL in relation to plates under darkness was observed secreted by P. digitatum (Macarisin et al., 2007; Ballester et al.,
over eight days. 2006) as an infection strategy.
Studies concerning the citrusePenicillium system have mainly
focused on new alternatives for controlling the disease or on the 2.3. Organic acid secretion by the pathogen for pH modulation
analysis of the mechanisms involved in induced resistance to fungal
infection. However, the recently sequenced complete genome An interesting strategy used by the pathogen to infect the host is
opened up new possibilities for investigating the virulence factors the secretion of organic acids resulting in pH modulation. Con-
related to the host-pathogen interaction. Here, we describe the trolling the pH has been considered a factor and a regulatory cue
known mechanisms and important genes related to cit- used by P. expansum, P. digitatum, and P. italicum for increased
rusePenicillium interactions. Understanding the infection process pathogenicity. The authors verified a pH difference of 1.65 between
and fungal strategies may represent an important step in devel- the healthy oranges (pH 4.77 ± 0.45) and decayed oranges (pH
oping ways to protect citrus from P. digitatum infections, leading to 3.12 ± 0.07) infected by P. digitatum. Prusky et al. (2004) suggested
more productive citriculture worldwide. that P. digitatum could enhance its pathogenicity through local pH
modulation of hosts, leading to an optimal pH for specific cell wall-
2. P. digitatum e citrus interaction degrading enzymes, such as polygalacturonases (PG) (Barmore and
Brown, 1981; Prusky et al., 2004; Zhang et al., 2013).
There are several factors that mediate and affect the interaction
between P. digitatum and its host during infection. In this section, 2.4. Flavonoid production in citrus fruits as a natural defense
we briefly describe the molecules and variables that directly or barrier
indirectly influence the infection process.
Another factor that influences the P. digitatum-citrus interaction
2.1. Hydrogen peroxide production as a plant response is the production of flavonoids by the host as defense barriers.
Flavonoids are well-documented as part of plant defense in Citrus
Hydrogen peroxide production by the host has been described spp. as reported by different studies (Ortun ~ o and Del Río, 2009;
as an important defense mechanism, which is bypassed by Ortun~ o et al., 2011; Kim et al., 2011). Flavonoids act as a chemical
P. digitatum due to increased catalase production. Hydrogen barrier in the outermost citric tissue and can be considered phy-
peroxide is mostly an initial consequence of plant defense in toalexins against P. digitatum, such as the polymethoxyflavones
response to pathogen infection, functioning as a signaling molecule localized at the flavedo level of fruits and the flavanones localized at
(Levine et al., 1994; Mellersh et al., 2002; Qin et al., 2011), and it has the albedo level (Fig. 1) (Ortun ~ o and Del Río, 2009; Ortun~ o et al.,
been reported to play a role as a signal in the induction of systemic 2011). Studies indicate changes in the levels of flavanone glyco-
acquired resistance (SAR) in plants (Neuencchwander et al., 1995; side - hesperidin (decreased approximately 7.8 %) and its corre-
Hunt et al., 1996; Conrath, 2006). Hydrogen peroxide also partici- sponding aglycon - hesperetin (increased approximately 7.2 %) at
pates in lignification processes of the cell wall, oxidative cross- five days after inoculation with P. digitatum. Similar results were
linking of proteins (Olson and Varner, 1993; Brisson et al., 1994), also found in lemons and grapefruits and can be associated with the
phytoalexin biosynthesis (Apostol et al., 1989) and induction of host hydrolyzing action of P. digitatum (Ortun ~ o and Del Río, 2009;
defense genes (Nathues et al., 2004), inhibiting pathogen growth Ortun~ o et al., 2011).
(Lu and Higgins, 1999; Ku zniak and Urbanek, 2000; García-Olmedo Ortun~ o et al. (2006) described polymethoxyflavones 5,6,7,30 ,40 -
et al., 2001). pentamethoxyflavone (sinensetin), 5,6,7,8,30 ,40 -hexamethoxy-
Macarisin et al. (2007) performed a detailed study on the level of flavone (nobiletin), 3,5,6,7,8,30 ,40 -heptamethoxyflavone and
hydrogen peroxide during the infection process of citrus fruits with 5,6,7,8,40 -pentamethoxyflavone (tangeretin) as important com-
P. digitatum (compatible interactions) and with Penicillium expan- pounds in citrus defense against pathogens. Furthermore, these
sum (nonhost interactions). The authors presented clear evidence researchers observed in vitro that nobiletin followed by hesperidin
that P. digitatum is able to avoid the H2O2-oxidative burst during the and naringin were the flavonoids that exhibited higher activities
first 25 h after inoculation in host cells, while P. expansum triggers a 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 2.6. Production of natural products by the fungi and pathogenicity
(Ortun ~ o et al., 2011). Interestingly, the corresponding aglycones are
more active as fungistatic agents against P. digitatum than the gly- There is evidence that several phytopathogenic fungi have the
cosylated flavanones (Ortun ~ o and Del Río, 2009), which is in ability to produce small molecules able to decrease plant defense
contrast to the hydrolyzing action of P. digitatum and requires responses, causing necrotic reactions in the infected cells
further investigation. (Morrissey and Osbourn, 1999). For instance, the virulence of
In addition, citrus defense mechanisms studies against the several fungi (Cochliobolus heterostrophus, C. miyabeanus, Fusarium
phytopathogen P. digitatum demonstrated that scoparone, an graminearum and Alternaria brassicicola) on their respective host
antifungal phenolic compound derived from phenylpropanoid plants has been reported to be mediated by particular siderophores,
metabolism, plays an important role as a citric phytoalexin present a class of secondary metabolites involved in iron uptake (Fox and
in oil glands (Kim et al., 1991; Ben-Yehoshua et al., 2008). Other Howlett, 2008).
studies on the role of flavonoids in the defense mechanisms of In this context, during P. digitatum infection, Ariza et al. (2002)
citrus fruits against the P. digitatum can be checked in the following identified secondary metabolites produced in P. digitatum
references (Del Río et al., 1998, 2004; Kim et al., 2011; Ballester biomass: the indole alkaloids tryptoquialanine A (1) and trypto-
et al., 2013). 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
2.5. Metabolomic and transcriptomic analysis of citrus fruits eburicol (7), with (1) and (2) being reported as major secondary
infected by P. digitatum metabolites in P. digitatum (Fig. 2). Secondary metabolites produced
by various microorganisms can contribute to the pathogenicity of
Tang et al. (2018) performed a search for the mechanism of several pathogenic fungi (Scharf et al., 2014). However, trypto-
deterioration of Powell orange pulps from Citrus sinensis infected by quialanine A (1), tryptoquialanine B (2) and other P. digitatum
P. digitatum after 40 h and 60 h using GCeMS to determine the secondary metabolites have not been directly related to the path-
levels of 26 polar primary metabolites. The authors also used the ogenicity of P. digitatum (Zhu et al., 2017); therefore, the exact
headspaced solid phase microextraction and the gas- biological role of these secondary metabolites has not been deter-
chromatography-mass spectrometry (HS-SPME-GC-MS) technique mined to date.
for the determination of 48 volatile organic compounds (VOCs). The Attempting to investigate the biological role of tryptoquialanine
authors verified a change in the levels of these VOCs, similar to A (1), Costa et al. (2019) recently analyzed the surface of oranges
octanoic acid ethyl ester, ethanol and aldehydes, which increased with green mold disease through mass spectrometry imaging and
the level in citrus fruits infected, suggesting that there is metabolic observed that tryptoquialanines are the major metabolites pro-
reprogramming during P. digitatum infection. It was also observed duced during the decay process on the orange surface. Costa et al.
that jasmonates and ethylene pathways are involved in the plant (2019) performed insecticidal bioassays using tryptoquialanine A
defense response for this interaction, as reported previously in and observed high toxicity against Aedes aegypti larvae. Larvae
other pathogen-host interactions (Tang et al., 2018). In addition, exposed to tryptoquialanine A presented a mortality rate of 37 %
Tang et al. (2018) performed a transcriptomic analysis of within 24 h and of 81 % within 96 h, indicating an important
P. digitatum infection and verified that there was activation of some insecticidal action during the infection process. These insecticidal
transcription factors: 29 MYB, 25 WRKY and 33 AP2/ERF were assays suggest that the tryptoquialanines are involved in the pro-
upregulated in fruits after 60 h of infection, suggesting their tection of the pathogen and the rotten citrus against insects, which
involvement in the fruits' defense against the pathogen (Tang et al., explains the production of the tryptoquialanines by P. digitatum on
2018). 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 great potential for secondary metabolite production. However,
indole alkaloids involved in the P. digitatum-citrus interaction, notably few studies have been able to correlate these metabolites to
suggesting that tryptoquialanine A is an effective biocontrol agent a biological role in the infection process.
against insects during orange decay. During this process, the volatiles emitted from the ruptured oil
In addition, Costa et al. (2019) reported for the first time for glands in wounded peel tissue can facilitate infection by promoting
P. digitatum the production of a new tryptoquialanine, tryptoquia- spore germination and germ tube elongation of P. digitatum (Droby
lanine C (8), and of intermediates involved in the tryptoquialanine et al., 2008).
A biosynthetic pathway, such as tryptoquialanone (9), 15-dimethyl- Ariza et al. (2002) aimed to understand the citrus-pathogen
2-epi-fumiquinazoline A (10), and deoxytryptoquialanone (11). interaction by comparing the volatile profile of healthy oranges,
Other secondary metabolites, such as tryptoquivaline L (12), tryp- oranges submitted to mechanical damage and oranges infected
toquivaline Q (13), fumiquinazoline A (14) and fumiquinazoline C with P. digitatum. These researchers observed higher amounts of
(15), have also been reported (Costa et al., 2019). limonene and other known citrus monoterpenes that were released
The discovery of metabolites produced by P. digitatum and the (instead of sesquiterpenes typical of healthy fruits) in mechanically
understanding of their biological role can provide more informa- injured fruits than healthy ones. Other studies concerning the role
tion on the infection process and how this fungus may survive to of limonene in the infection process and spore germination can be
environmental pressures in the host. Once the P. digitatum genome checked in the following references (Rodríguez et al., 2015; Tao
has been entirely sequenced, there is abundant information on the et al., 2019). Regarding oranges contaminated with P. digitatum,
22
J.H. Costa et al. / Fungal Biology 123 (2019) 584e593
understand citrus colonization by P. digitatum since the whole of glucose-repressible genes, controlling the use of carbon sources
mechanism of infection has not been determined to date (Vu et al., and, in the case of plant pathogenic fungi, the expression of CWDE
2018; Zhu et al., 2017; Zhang et al., 2013b). (Zhang et al., 2013c; Nadal et al., 2010; Yi et al., 2008). Concerning
virulence, Zhang et al. (2013c) observed that the DPdSFN1 mutants
3.1. Transporters were still pathogenic to citrus, but the symptoms took much longer
to develop in the fruits, and the production of conidia was affected.
The putative sucrose uptake transporter (SUT) gene PdSUT1 was Expression of the CWDE genes was upregulated for the wild-type
disrupted in P. digitatum (Ramo n-Carbonell and Sa nchez-Torres, when induced with pectin, but in the DPdSNF1 mutants, the
2017b). Initially, the SUTs were considered specific to plants; expression levels almost did not change (Zhang et al., 2013c). Zhang
however, Schizosaccharomyces pombe was the first fungus found et al. (2013c) showed that the PdSNF1 gene is involved in the
with the homologous gene Sut1p (Ramo n-Carbonell and Sa nchez- regulation of CWDE gene expression in P. digitatum and, conse-
Torres, 2017b; Reinders and Ward, 2001). SUTs have also been re- quently, in virulence.
ported in other fungi and related to virulence, but the main func- CWDEs are also affected in P. digitatum by the mitogen-activated
tion and regulation of fungal sucrose uptake transporters have not protein kinase B gene PdMPkB. Through qRT-PCR analyses, Ma et al.
been thoroughly elucidated to date (Ramo n-Carbonell and (2016) verified that CWDE genes were downregulated in DPdMPkB
Sanchez-Torres, 2017b). DPdSUT1 mutants had lower virulence mutants. Therefore, in fruits, DPdMPkB mutants showed fewer le-
than the wild type in citrus infection, while mutants with over- sions than the wild type (Ma et al., 2016). The effect of pH signaling
expression in PdSUT1 did not present an increase in virulence transcription factor gene PdpacC deletion in P. digitatum was stud-
(Ramo n-Carbonell and Sa nchez-Torres, 2017b). ied by Zhang et al. (2013a), and the disruption of this gene also
Expression of the genes involved in the major facilitator su- leads to fewer virulence mutants since the expression levels of the
perfamily (MFS) transporters was also analyzed by Ramo n- polygalacturonase gene Pdpg2 and pectin lyase gene Pdpnl1 were
Carbonell and Sa nchez-Torres (2017b) and revealed that this not upregulated or were weakly upregulated compared to the wild-
expression is enhanced in the DPdSUT1 mutants, which may type strain. These results clearly reinforce that CWDEs are essential
explain the increase in resistance to fungicides since the MFS car- for P. digitatum virulence as for other phytopathogens. Additionally,
riers are powered and capable of carrying a wide variety of com- genes that are involved in CWDE gene regulation are consequently
pounds, such as toxic products and drugs (Ramo n-Carbonell and involved in pathogenicity.
Sanchez-Torres, 2017b, Law et al., 2008; Reddy et al., 2012).
Ramo n-Carbonell and Sa nchez-Torres (2017b) concluded that the 3.3. Signaling pathway components
PdSUT1 gene is not directly involved in virulence and resistance to
fungicides in P. digitatum. Instead, this gene only contributes to Studies on some signaling pathway genes demonstrated that
these factors because it activates the expression of the MFS trans- they are required for virulence but not directly involved in the
porter genes. pathogenic mechanisms of P. digitatum. For example, PdpacC is a
The ATP-binding cassette (ABC) and MFS transporters are signaling pathway gene that is essential for virulence because it
involved in virulence by mediating the secretion of host-specific regulates CWDE gene expression (Zhang et al., 2013a).
toxins or providing protection against plant defense components Another example is the gene Pdos2, which encodes the mitogen-
(Stergiopoulos et al., 2002). ABC transporter genes in P. digitatum activated protein kinase Hog1 studied by Wang et al. (2014) and is
are conserved and directly involved in drug resistance (Sa nchez- part of the high-osmolarity glycerol pathway. DPdos2 mutants are
Torrez and Tuset, 2011; Sun et al., 2013). In contrast, two MFS more sensitive to hyperosmotic stress and cell wall-disturbing
transporters were studied and related to fungicide resistance, agents and are more resistant to fludioxonil once this fungicide
although there are more than one hundred MFS genes in acts on the HOG pathway (Wang et al., 2014). Wang et al. (2014)
P. digitatum (Sun et al., 2013). Disruption of the MFS genes PdMfs1 concluded that Pdos2 is associated with positive regulation of
(Wang et al., 2012) and PdMfs2 (Wu et al., 2016) in P. digitatum leads glycerol synthesis and negative regulation of ergosterol synthesis,
to mutants that are more sensitive to fungicides and less virulent in and the decrease in virulence may result from the defect in vege-
citrus fruits. tative growth and sensitivity to osmotic stress.
The mechanism by which MFS plays a role during P. digitatum The role of the calcineurin-responsive transcription factor gene
infection requires further investigation (Wang et al., 2012). One of PdCrz1 in P. digitatum was investigated by Zhang et al. (2013b).
the possibilities is that a toxin transported by MFS acts as a viru- DPdCrz1 mutants were observed to be defective in conidiation,
lence factor (Wang et al., 2012) or that MFS plays a role in infection virulence, and hypersensitivity to stress caused by Ca2þ and DMI
through a different way. For example, Liu et al. (2017) recently fungicides. In addition, the expression of cell wall synthase genes
described that the ChMfs1 gene of the phytopathogenic fungus (CHS2, CHS3 and FKS1) leads to defective cell wall integrity (Zhang
Colletotrichum higginsianum is involved in intrahyphal hyphae for- et al., 2013). However, the mechanism by which PdCrz1 controls
mation, demonstrating a novel function of MFS transporters that P. digitatum virulence requires further investigation (Zhang et al.,
plays a key role in infection. 2013).
Asexual spores (conidia) are the inocula of green mold disease
3.2. Cell wall-degrading enzymes produced by P. digitatum during fruit-pathogen interaction, and
they play an important role in the disease cycle since germination
Phytopathogenic fungi need to pass through the plant cell wall, of conidia occurs in the surface wounds of citrus (Zhang et al.,
an important barrier, to attack. Thus, phytopathogenic fungi pro- 2013c; Ramo n-Carbonell and Sanchez-Torres, 2017a). The PdSNF1
duce cell wall-degrading enzymes (CWDEs) to depolymerize plant gene, as mentioned before, is required for P. digitatum conidiation
cell wall structures, such as cellulose and pectin (Kubicek et al., (Zhang et al., 2013c). PdSNF1 plays a role in the expression of PdBrlA
2014). The ability to degrade the host cell wall has long been and FadA, which are genes involved in conidiation and growth
thought to play a key role in P. digitatum infection such that CWDEs signaling (Zhang et al., 2013c). PdMPkB and PdSlt2 are other genes
are upregulated during the infection process (Zhang et al., 2013c). also involved in conidial formation (Ma et al., 2016; Ramo n-
Zhang et al. (2013c) verified the importance of the PdSNF1 gene Carbonell and Sa nchez-Torres, 2017c). Thus, DPdSNF1, DPdMPkB
for virulence. In microorganisms, SNF1 regulates the derepression and DPdSlt2 mutants showed less virulence compared to the wild-
23
J.H. Costa et al. / Fungal Biology 123 (2019) 584e593
Table 1
Main genes studied and their role in P. digitatum metabolism.
Transporters 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)
CWDE Pdpg2 Involved in virulence Zhang et al. (2013a)
Signal Pathways 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
Structural Components 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
Dispensable for virulence 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)
24
J.H. Costa et al. / Fungal Biology 123 (2019) 584e593
Sterol regulatory element-binding protein genes (Pdsre) were metabolism of P. digitatum and the roles of natural products in the
evaluated by Ruan et al. (2017) using three different P. digitatum infection merit further research. Innumerous biosynthetic gene
mutants. The lack mutants DPdsreA, DPdsreB and DPdsreAB caused clusters are coded in this phytopathogen's genome, and many gene
macerations of similar size to the lesions induced on mandarins by clusters are described to be upregulated during the infection pro-
the wild-type strain. Pdsre are principally involved in resistance to cess; however, only tryptoquialanine-like metabolites are
DMI fungicides because DPdsre mutants increased sensitivity to described as natural products for this fungus, meaning that many
these fungicides (Ruan et al., 2017). products remain to be discovered.
The mutants with deletions in tqaA, a gene involved in the
production of tryptoquialanines, showed no growth, development, Acknowledgments
stress response or pathogenicity variations, as reported by Zhu et al.
(2017), indicating that tryptoquialanines are not involved in the This study was financed in part by the Coordenaç~ ao de Aper-
interaction of P. digitatum with citrus host. Zhu et al. (2017) sug- feiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance
gested that tryptoquialanines may be involved in the protection of Code 001 and Fundaça ~o de Amparo a Pesquisa no Estado de Sa ~o
rotten citrus from insects. This hypothesis was confirmed, as Paulo [grant number FAPESP 2017/24462-4 and 2018/03670-0]. We
mentioned before by Costa et al. (2019), through the high insecti- also thank Eder Vilhena Araujo for the orange photos used in Fig. 1.
cide activity of tryptoquialanine A against A. aegypti larvae.
Similar results were also obtained for the deletion of PdKu80, a
References
gene of the nonhomologous end-joining (NHEJ) pathway (Xu et al.,
2014), and for ndo1, a gene coding a naphthalene dioxygenase Aguirre, J., Hansberg, W., Navarro, R., 2006. Fungal responses to reactive oxygen
(Lo pez-Perez et al., 2015); none of these genes was found to be species. Med. Mycol. 44, 101e107.
important for virulence. 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.
Table 1 summarizes the main genes studied for P. digitatum Ariza, M.R., Larsen, T.O., Petersen, B.O., Duus, J.Ø., Barrero, A.F., 2002. Penicillium
found in the literature. digitatum metabolites on synthetic media and citrus fruits. J. Agric. Food Chem.
The functions of most of the studied P. digitatum genes are 50, 6361e6365.
Ballester, A.R., Lafuente, M.T., Gonza lez-Candelas, L., 2006. Spatial study of antiox-
conserved across the species. The genes that affect virulence are idant enzymes, peroxidase and phenylalanine ammonia-lyase in the citrus
generally related to growth, asexual reproduction, cell wall integ- fruitePenicillium digitatum interaction. Postharvest Biol. Technol. 39, 115e124.
rity or the expression of other genes (CWDEs or transporters). Ballester, A.R., Lafuente, M.T., Gonzalez-Candelas, L., 2013. Citrus phenylpropanoids
and defence against pathogens. Part II: gene expression and metabolite accu-
Some genes, such as PdCrz1, have not had their role in infection
mulation in the response of fruits to Penicillium digitatum infection. Food Chem.
thoroughly elucidated and warrant further investigation. As 136, 285e291.
mentioned previously, the functions of the transporters on viru- Barmore, C.R., Brown, G.E., 1981. Polygalacturonase from citrus fruit infected with
lence have not been well-established, as no secondary metabolites Penicillium italicum. Phytopathology 71, 328e331.
Ben-Yehoshua, S., Rodov, V., Nafussi, B., Feng, X., Yen, J., Koltai, T., Nelkenbaum, U.,
have been linked to the infection process, representing a gap in the 2008. Involvement of limonene hydroperoxides formed after oil gland injury in
research concerning the fungus P. digitatum. The studies reinforced the induction of defense response against Penicillium digitatum in lemon fruit.
that cell wall-degrading enzymes seem to be involved in the viru- J. Agric. Food Chem. 56, 1889e1895.
Brisson, L.F., Tenhaken, R., Lamb, C., 1994. Function of oxidative crosslinking of cell
lence of this fungus and help to explain how P. digitatum modulates wall structural proteins in plant disease resistance. Plant Cell 6, 1703e1712.
the acidity of the environment. However, the molecular biology of Chaga, G.S., Medin, A.S., Chaga, S.G., Porath, J.O., 1992. Isolation and characterization
P. digitatum warrants a thorough investigation, and an extensive of catalase from Penicillium chvysogenum. J. Chromatogr A. 604, 177e183.
Citrus: World Markets and Trade U.S. Department of Agriculture, Washington, DC.
analysis of the gene clusters found in the genome of this important https://apps.fas.usda.gov/psdonline/circulars/citrus.pdf (accessed 16 May 2019).
pathogen is necessary to better understand the entire infection Conrath, U., 2006. Systemic acquired resistance. Plant Signal. Behav. 1, 179e184.
process. 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
4. Conclusions 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,
Despite the great economic interest in P. digitatum, the molec-
Penicillium digitatum, and Geotrichum species. J. Agric. Food Chem. 46,
ular basis of infection and specificity towards the citrus host re- 4423e4428.
mains largely unknown. In recent years, most of the research on Del Río, J.A., Go mez, P., Baidez, A.G., Arcas, M.C., Botía, J.M., Ortun ~ o, A., 2004.
P. digitatum has focused on treatments against infection symptoms, Changes in the levels of polymethoxyflavones and flavanones as part of the
defense mechanism of Citrus sinensis (Cv. Valencia Late) fruits against Phy-
fungicide resistance and the biocontrol of this fungus using tophthora citrophthora. J. Agric. Food Chem. 52, 1913e1917.
antagonist microorganisms. However, since 2012, with the publi- Droby, S., Eick, A., Macarisin, D., Cohen, L., Rafael, G., Stange, R., McColum, G.,
cation of the genome of this important necrotrophic fungus, this Dudai, N., Nasser, A., Wisniewski, M., Shapira, R., 2008. Role of citrus volatiles in
host recognition, germination and growth of Penicillium digitatum and Penicil-
scenario has been changing, and many genes have been determined lium italicum. Postharvest Biol. Technol. 49 (3), 386e396.
to be virulence factors and to be important in the disease. The Fox, E.M., Howlett, B.J., 2008. Secondary metabolism: regulation and role in fungal
knowledge concerning the genes required for full virulence biology. Curr. Opin. Microbiol. 11 (6), 481e487.
Frisvad, J.C., Samson, R.A., 2004. Polyphasic taxonomy of Penicillium subgenus
reviewed in this study (PdSNF1, PdMPkB, Pdos2, PdSte12, PdCrz1, Penicillium A guide to identification of food and air-borne terverticillate Peni-
Pdpmt2, PdchsVII) provides the first insights about pathogen-host cillia and their mycotoxins. Stud. Mycol. 49, 1e174.
interaction in P. digitatum and provides initial thoughts on poten- Gandía, M., Harries, E., Marcos, J.F., 2012. Identification and characterization of
chitin synthase genes in the postharvest citrus fruit pathogen Penicillium dig-
tial target fungicides to be developed against such phytopathogens. itatum. Fungal Biol. 116, 654e664.
The main infection strategies of P. digitatum, citrus responses, sec- Gandía, M., Harries, E., Marcos, J.F., 2014. The myosin motor domain-containing
ondary metabolites and genes involved in host-pathogen interac- chitin synthase PdChsVII is required for development, cell wall integrity and
virulence in the citrus postharvest pathogen Penicillium digitatum. Fungal
tion are represented in Fig. 3. In our opinion, understanding the
Genet. Biol. 67, 58e70.
infection process and the defense responses of the hosts, as well as García-Olmedo, F., Rodríguez-Palenzuela, P., Molina, A., Alamillo, J.M., Lo pez-
the virulence mechanisms of this pathogen, is the first step in Solanilla, E., Berrocal-Lobo, M., 2001. Antibiotic activities of peptides, hydrogen
developing safer and more eco-friendly alternative strategies for peroxide and peroxynitrite in plant defence. FEBS Lett. 498, 219e222.
Ghooshkhaneh, N.G., Golzarian, M.R., Mamarabadi, M., 2018. Detection and classi-
controlling citrus postharvest diseases, leading to new and safer fication of citrus green mold caused by Penicillium digitatum using multispectral
control strategies for green mold disease. In addition, the secondary imaging. J. Sci. Food Agric. 98 (9), 3542e3550.
25
Hao, W., Li, H., Hu, M., Yang, L., Rizwan-ul-Haq, M., 2011. Integrated control of citrus Neuencchwander, U., Vernooij, B., Friedrich, L., Uknes, S., Kessmann, H., Ryals, J.,
green and blue mold and sour rot by Bacillus amyloliquefaciens in combination 1995. Is hydrogen peroxide a second messenger of salicylic acid in systemic
with tea saponin. Postharvest Biol. Technol. 59, 316e323. acquired resistance? Plant J. 8, 227e233.
Harries, E., Gandía, M., Carmona, L., Marcos, J.F., 2015. The Penicillium digitatum Nimrichter, L., Rodrigues, M.L., Barreto-Bergter, E., Travassos, L.R., 2008. Sophisti-
protein O-mannosyltransferase Pmt2 is required for cell wall integrity, con- cated functions for a simple molecule: the role of glucosylceramides in fungal
idiogenesis, virulence and sensitivity to the antifungal peptide PAF26. Mol. cells. Lipid Insights 2, 61e73.
Plant Pathol. 16, 748e761. Olson, P.D., Varner, J.E., 1993. Hydrogen peroxide and lignification. Plant J. 4,
Hunt, M.D., Neuenschwander, U.H., Delaney, T.P., Weymann, K.B., Friedrich, L.B., 887e892.
Lawton, K.A., Steiner, H.Y., Ryals, J.A., 1996. Recent advances in systemic ac- Ortun~ o, A., Del Río, J.A., 2009. Role of citrus phenolic compounds in the resistance
quired resistance research- a review. Gene 179, 89e95. mechanism against pathogenic fungi. Tree For. Sci. Biotech. 3, 49e53.
Ismail, M.A., Zhang, J., 2004. Post-harvest citrus diseases and their control. Outlooks Ortun~ o, A., B
aidez, A., Go mez, P., Arcas, M.C., Porras, I., García-Lido n, A., Del Río, J.A.,
Pest Manag. 15, 29e35. 2006. Citrus paradisi and Citrus sinensis flavonoids: their influence in the
Julca, I., Droby, S., Sela, N., Marcet-Houben, M., Gabaldo, T., 2015. Contrasting defence mechanism against Penicillium digitatum. Food Chem. 98, 351e358.
genomic diversity in two closely related postharvest pathogens: Penicillium Ortun~ o, A., Díaz, L., Alvarez, N., Porras, I., García-Lido n, A., Del Río, J.A., 2011.
digitatum and Penicillium expansum. Genome. Biol. Evol. 8, 218e227. Comparative study of flavonoid and scoparone accumulation in different Citrus
Kanetis, L., Fo €rster, H., Adaskaveg, J.E., 2010. Determination of natural resistance species and their susceptibility to Penicillium digitatum. Food Chem. 125, 232e239.
frequencies in Penicillium digitatum using a new air-sampling method and Parveen, S., Wani, A.H., Bhat, M.Y., Koka, J.A., 2016. Biological control of postharvest
characterization of Fludioxonil- and Pyrimethanil-Resistant isolates. Phytopa- fungal rots of rosaceous fruits using microbial antagonists and plant extracts e
thology 100, 738e743. a review. Czech Mycol. 68, 41e46.
Kim, J.J., Ben-Yehoshua, S., Shapiro, B., Henis, Y., Carmeli, S., 1991. Accumulation of Perez, M.F., Ibarreche, J.P., Isas, A.S., Sepulveda, M., Ramallo, J., Dib, J.R., 2017.
scoparone in heat-treated lemon fruit inoculated with Penicillium digitatum Antagonistic yeasts for the biological control of Penicillium digitatum on lemons
Sacc. Plant Physiol. 97, 880e885. stored under export conditions. Biol. Control 115, 135e140.
Kim, H.G., Kim, G.S., Lee, J.H., Park, S., Jeong, W.Y., Kim, J.H., Kin, S.T., Cho, Y.A., Poppe, L., Vanhoutte, S., Ho €fte, M., 2003. Modes of action of Pantoea agglomerans
Lee, W.S., Lee, S.J., Shin, S.C., 2011. Determination of the change of flavonoid CPA-2, an antagonist of postharvest pathogens on fruits. Eur. J. Plant Pathol. 109,
components as the defense materials of Citrus unshiu Marc. fruit peel against 963e973.
Penicillium digitatum by liquid chromatography coupled with tandem mass Prusky, D., McEvoy, J.L., Saftner, R., Conway, W.S., Jones, R., 2004. Relationship be-
spectrometry. Food Chem. 128, 49e54. tween host acidification and virulence of Penicillium spp. on apple and citrus
Kubicek, C.P., Starr, T.L., Glass, N.L., 2014. Plant cell wall-degrading enzymes and fruit. Phytopathology 94, 44e51.
their secretion in plant-pathogenic fungi. Annu. Rev. Phytopathol. 52, Qin, G., Liu, J., Cao, B., Li, B., Tian, S., 2011. Hydrogen peroxide acts on sensitive
427e451. mitochondrial proteins to induce death of a fungal pathogen revealed by pro-
Kuzniak, E., Urbanek, H., 2000. The involvement of hydrogen peroxide in plant teomic analysis. PLoS One 6, e21945.
responses to stresses. Acta Physiol. Plant. 22, 195e203. Ramo n-Carbonell, M., Sa nchez-Torres, P., 2017a. The transcription factor PdSte12
Ladanyia, M., 2010. Citrus Fruit: Biology, Technology and Evaluation, first ed. Aca- contributes to Penicillium digitatum virulence during citrus fruit infection.
demic Press. ISBN-10: 0123741300. Postharvest Biol. Technol. 125, 129e139.
Lafuente, M.T., Alfe rez, F., Gonza lez-Candelas, L., 2018. Light-emitting diode blue Ramo n-Carbonell, M., Sa nchez-Torres, P., 2017b. Involvement of Penicillium dig-
light alters the ability of Penicillium digitatum to infect citrus fruits. Photochem. itatum PdSUT1 in fungicide sensitivity and virulence during citrus fruit infec-
Photobiol. 94, 1003e1009. tion. Microbiol. Res. 203, 57e67.
Law, C.J., Maloney, P.C., Wang, D., 2008. Ins and outs of major facilitator superfamily Ramo n-Carbonell, M., Sa nchez-Torres, P., 2017c. PdSlt2 Penicillium digitatum
antiporters. Annu. Rev. Microbiol. 62, 289e305. mitogen activated-protein kinase controls sporulation and virulence during
Levine, A., Tenhaken, R., Dixon, R., Lamb, C., 1994. H2O2 from the oxidative burst citrus fruit infection. Fungal Biol. 121, 1063e1074.
orchestrates the plant hypersensitive disease resistance response. Cell 79, Reddy, V.S., Shlykov, M.A., Castillo, R., Sun, E.I., Saier, M.H., 2012. The Major Facili-
583e593. tator Superfamily (MFS) Revisited,. FEBS J. 279 (11), 2022e2035.
Liu, L., Yan, Y., Huang, J., Hsiang, T., Wei, Y., Li, Y., Gao, J., Zheng, L., 2017. A novel MFS Reinders, A., Ward, J.M., 2001. Functional characterization of the a-glucoside
transporter gene ChMfs1 is important for hyphal morphology, conidiation, and transporter Sut1p from Schizosaccharomyces pombe, the first fungal homologue
pathogenicity in Colletotrichum higginsianum. Front. Microbiol. 8, 1953. of plant sucrose transporters. Mol. Microbiol. 39, 445e454.
Lopez-Pe rez, M., Ballester, A.R., Gonz alez-Candelas, L., 2015. Identification and Rittershaus, P.C., Kechichian, T.B., Allegood, J.C., Merrill, A.H., Hennig, M., Luberto, C.,
functional analysis of Penicillium digitatum genes putatively involved in viru- Del Poeta, M., 2006. Glucosylceramide synthase is an essential regulator of
lence towards citrus fruit. Mol. Plant Pathol. 16, 262e275. pathogenicity of Cryptococcus neoformans. J. Clin. Investig. 116, 1651e1659.
Louw, J.P., Korsten, L., 2019. Impact of ripeness on the infection and colonisation of Rodríguez, A., Shimada, T., Cervera, M., Redondo, A., Alque zar, B., Rodrigo, M.J.,
Penicillium digitatum and P. expansum on plum. Postharvest Biol. Technol. 149, Zacarías, L., Palou, L., Lo pez, M.M., Pen ~ a, L., 2015. Resistance to pathogens in
148e158. terpene down-regulated orange fruits inversely correlates with the accumula-
Lu, H., Higgins, V.J., 1999. The effect of hydrogen peroxide on the viability of tomato tion of D-limonene in peel oil glands. Plant Signal. Behav. 10 (6) e1028704-
cells and of the fungal pathogen Cladosporium fulvum. Physiol. Mol. Plant 1ee1028704-4.
Pathol. 54, 131e143. Ruan, R., Wang, M., Liu, X., Sun, X., Chung, K.R., Li, H., 2017. Functional analysis of
Ma, H., Sun, X., Wang, M., Gai, Y., Chung, K.R., Li, H., 2016. The citrus postharvest two sterol regulatory element binding proteins in Penicillium digitatum. PLoS
pathogen Penicillium digitatum depends on the PdMpkB kinase for develop- One 12 (5), e0176485.
mental and virulence functions. Int. J. Food Microbiol. 236, 167e176. Ruiz-Herrera, J., Gonza lez-Prieto, J.M., Ruiz-Medrano, R., 2002. Evolution and
Macarisin, D., Cohen, L., Eick, A., Rafael, G., Belausov, E., Wisniewski, M., Droby, S., phylogenetic relationships of chitin synthases from yeasts and fungi. FEMS
2007. Penicillium digitatum suppresses production of hydrogen peroxide in host Yeast Res. 1, 247e256.
tissue during infection of citrus fruit. Phytopathology 97, 1491e1500. Sa
nchez-Torrez, P., Tuset, J.J., 2011. Molecular insights into fungicide resistance in
Marcet-Houben, M., Ballester, A.R., la Fuente, B., Harries, E., Marcos, J.F., Gonz alez- sensitive and resistant Penicillium digitatum strains infecting citrus. Postharvest
Candelas, L., Galbado n, T., 2012. Genome sequence of the necrotrophic fungus Biol. Technol. 59, 159e165.
Penicillium digitatum, the main postharvest pathogen of citrus. BMC Genomics Scharf, D.H., Heinekamp, T., Brakhage, A.A., 2014. Human and plant fungal patho-
13, 646. gens: the role of secondary metabolites. PLoS Pathog. 10 (1), e1003859.
Mellersh, D.G., Foulds, I.V., Higgins, V.J., Heath, M.C., 2002. H2O2 plays different roles Stergiopoulos, I., Zwiers, L., De Waard, M.A., 2002. Secretion of natural and synthetic
in determining penetration failure in three diverse plant-fungal interactions. toxic compounds from filamentous fungi by membrane transporters of the ATP-
Plant J. 29, 257e268. binding cassette and major facilitator superfamily. Eur. J. Plant Pathol. 108 (7),
Molto , E., Blasco, J., Go
mez-Sanchıs, J., 2010. Analysis of hyperspectral images of 719e734.
citrus fruits. In: Sun, D.W. (Ed.), Hyperspectral Imaging for Food Quality Anal- Sun, X., Ruan, R., Lin, L., Zhu, C., Zhang, T., Wang, M., Li, H., Yu, D., 2013. Genomewide
ysis and Control. Academic Press, pp. 321e348. investigation into DNA elements and ABC transporters involved in imazalil
Morrissey, J.P., Osbourn, A.E., 1999. Fungal resistance to plant antibiotics as a resistance in Penicillium digitatum. FEMS Microbiol. Lett. 348, 11e18.
mechanism of pathogenesis. Microbiol. Mol. Biol. Rev. 63 (3), 708e724. Talibi, I., Boubaker, H., Boudyach, E.H., Aoumar, A.A.B., 2014. Alternative methods for
Mouyna, I., Kniemeyer, O., Jank, T., Loussert, C., Mellado, E., Aimanianda, V., the control of postharvest citrus diseases. J. Appl. Microbiol. 117, 1e17.
Beauvais, A., Wartenberg, D., Sarfati, J., Bayry, J., Pre vost, M.C., Brakhage, A.A., Tang, N., Chen, N., Hu, N., Deng, W., Chen, Z., Li, Z., 2018. Comparative metabolomics
Strahl, S., Huerre, M., Latge , J.P., 2010. Members of protein O-mannosyl- and transcriptomic profiling reveal the mechanism of fruit quality deterioration
transferase family in Aspergillus fumigatus differentially affect growth, and the resistance of citrus fruit against Penicillium digitatum. Postharvest Biol.
morphogenesis and viability. Mol. Microbiol. 76, 1205e1221. Technol. 145, 61e73.
Nadal, M., Garcia-Pedrajas, M.D., Gold, S.E., 2010. The snf1 Gene of ustilago maydis Tao, N., Chen, Y., Wu, Y., Wang, X., Li, L., Zhu, A., 2019. The terpene limonene induced
acts as a dual regulator of cell wall degrading enzymes. Phytopathology 100, the green mold of citrus fruit through regulation of reactive oxygen species
1364e1372. (ROS) homeostasis in Penicillium digitatum spores. Food Chem. 277, 414e422.
Nathues, E., Joshi, S., Tenberge, K.B., von den Driesch, M., Oeser, B., B€ aumer, N., Vilanova, L., Teixido , N., Torres, R., Usall, J., Vin
~ as, I., Sa
nchez-Torres, P., 2016. Rele-
Mihlan, M., Tudzynski, P., 2004. CPTF1, a CREB-like transcription factor, is vance of the transcription factor PdSte12 in Penicillium digitatum conidiation
involved in the oxidative stress response in the phytopathogen Claviceps pur- and virulence during citrus fruit infection. Int. J. Food Microbiol. 235, 93e102.
purea and modulates ROS level in its host Secale cereal. Phytopathology 17, 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.,
383e393. Tran, V.-T., 2018. A highly efficient Agrobacterium tumefaciens-mediated
26
J.H. Costa et al. / Fungal Biology 123 (2019) 584e593
transformation system for the postharvest pathogen Penicillium digitatum using Yi, M., Park, J.H., Ahn, J.H., Lee, Y.H., 2008. MoSNF1 regulates sporulation and
DsRed and GFP to visualize citrus host colonization. J. Microbiol. Methods 144, pathogenicity in the rice blast fungus Magnaporthe oryzae. Fungal Genet. Biol.
134e144. 45, 1172e1181.
Wang, J.Y., Li, H.Y., 2008. Agrobacterium tumefaciens-mediated genetic trans- Zhang, T., Sun, X., Xu, Q., Candelas, L.G., Li, H., 2013a. The pH signaling transcription
formation of the phytopathogenic fungus Penicillium digitatum. J. Zhejiang Univ. factor PacC is required for full virulence in Penicillium digitatum. Appl. Micro-
Sci. B 9, 823e828. biol. Biotechnol. 97, 9087e9098.
Wang, J., Sun, X., Lin, L., Zhang, T., Ma, Z., Li, H., 2012. PdMfs1, a major facilitator Zhang, T., Xu, Q., Sun, X., Li, H., 2013b. The calcineurin-responsive transcription
superfamily transporter from Penicillium digitatum, is partially involved in factor Crz1 is required for conidation, full virulence and DMI resistance in
the imazalil-resistance and pathogenicity. Afr. J. Microbiol. Res. 6 (1), Penicillium digitatum. Microbiol. Res. 168, 211e222.
95e105. Zhang, T., Sun, X., Xu, Q., Zhu, C., Li, Q., Li, H., 2013c. PdSNF1, a sucrose non-
Wang, M., Chen, C., Zhu, C., Sun, X., Ruan, R., Li, H., 2014. Os2 MAP kinase-mediated fermenting protein kinase gene, is required for Penicillium digitatum con-
osmostress tolerance in Penicillium digitatum is associated with its positive idiation and virulence. Appl. Microbiol. Biotechnol. 97, 5433e5445.
regulation on glycerol synthesis and negative regulation on ergosterol syn- Zhu, C., Wang, M., Wang, W., Ruan, R., Ma, H., Mao, C., 2014. Glucosylceramides are
thesis. Microbiol. Res. 169, 511e521. required for mycelial growth and full virulence in Penicillium digitatum. Bio-
Wang, W., Wang, M., Wang, J., Zhu, C., Chung, K., Li, H., 2016. Adenylyl cyclase is chem. Biophys. Res. Commun. 455, 165e171.
required for cAMP production, growth, conidial germination, and virulence in the Zhu, C., Wang, W., Wang, M., Ruan, R., Sun, X., He, M., Mao, C., Li, H., 2015. Deletion
citrus green mold pathogen Penicillium digitatum. Microbiol. Res. 192, 11e20. of PdMit1, a homolog of yeast Csg1, affects growth and Ca2þ sensitivity of the
Wani, S.H., 2010. Inducing fungus-resistance into plants through biotechnology. fungus Penicillium digitatum, but does not alter virulence. Res. Microbiol. 166,
Not. Sci. Biol. 2, 14e21. 143e152.
Wu, Z., Wang, S., Yuan, Y., Zhang, T., Liu, J., Liu, D., 2016. A novel facilitator super- Zhu, C., Sheng, D., Wu, X., Wang, M., Hu, X., Li, H., Yu, D., 2017. Identification of
family transporter in Penicillium digitatum (PdMFS2) is required for prochloraz secondary metabolite biosynthetic gene clusters associated with the infection
resistance, conidiation and full virulence. Biotechnol. Lett. 38 (1), 1349e1357. of citrus fruit by Penicillium digitatum. Postharvest Biol. Technol. 134, 17e21.
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.
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
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
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
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.
https://doi.org/10.1016/j.funbio.2019.03.002
1878-6146/© 2019 British Mycological Society. Published by Elsevier Ltd. All rights reserved.
30
J.H. Costa et al. / Fungal Biology 123 (2019) 594e600
compared to the wild type strain, so the exact biological role of 2.4. Isolation of secondary metabolites 1, 3 and 7
tryptoquialanines is still unknown (Zhu et al., 2017).
The studies concerning this pathogen-host system mainly focus High performance liquid chromatography (HPLC) separations
on the fruit response to the pathogen or on the fungicide resis- for 1, 3 and 7 were performed on a Phenomenex column Luna 5 mm
tance (Lope z-Perez et al., 2015), so the pathogenicity mechanisms Phenyl-Hexyl (250 4.6 mm) using a SHIMADZU prominence HPLC
and infection process are still unclear (Zhu et al., 2017). Therefore, LC-20AT, equipped with CBM-20A communication bus module,
it become necessary to seek a better understanding of these SPD-M20A photodiode array detector and SIL-20A auto sampler.
mechanisms in different citrus fruits, in order to discover sec- The mobile phase was water (A) and acetonitrile with 0.1 % (v/v) of
ondary metabolites potentially associated with the citrus infection. formic acid (B). Flow rate was 1.0 mL min 1. The eluent profile (A:B)
Understanding the infection process and the fungus strategies is was: 0e50 min, gradient from 65:35 to 50:50; 50e70 min, gradient
an important step to develop ways to protect citrus from from 50:50 to 40:60. Preparative HPLC purifications for 1, 3 and 7
P. digitatum infections leading to a more productive citriculture were performed on a Phenomenex column Luna 5 mm Phenyl-Hexyl
worldwide. (250 10 mm) using a Waters 1525 Binary HPLC Pump equipped
In this context, Mass Spectrometry Imaging (MSI) is a powerful with Waters 2998 Photodiode Array Detector and Waters Fraction
and useful tool to describe the functional roles of secondary me- Collector III using the same optimized gradient conditions with a
tabolites in a biological context, based on their relative abundances flow rate set at 4.7 mL min 1.
of m/z and spatial distribution (Lei et al., 2011). Through MSI it is
possible to investigate and identify secondary metabolites involved 2.5. Characterization of secondary metabolites 1, 3 and 7
at different stages of a phytopathogen infection. Here we explored
1
different P. digitatum infection stages by Mass Spectrometry Im- H NMR, 13C NMR and 2D experiments were acquired in a
aging combined with Molecular Networking using GNPS approach, Bruker Avance III 500 (1H 500.13 MHz and 13C 125.7 MHz).
to profile the secondary metabolite production of P. digitatum Deuterated chloroform (CDCl3; 7.23 ppm), dimethyl dulfoxide
during infection process. The biological role of indole alkaloids (DMSO; 2.50 ppm and 39.51 ppm) and tetramethylsilane (TMS;
during this process is also discussed. 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. Materials and methods
2.6. In vivo and in vitro assays
2.1. Fungal culture
For in vivo assays, mature oranges (Citrus sinensis), sicilian
The P. digitatum strain used in the studies is deposited with the lemons (Citrus limon) and tangerine (Citrus reticulata) obtained
Spanish Type Culture Collection (CECT) under the accession code from a local grocery store (Campinas, SP, Brazil) were surface-
CECT20796. P. digitatum was maintained on commercial potato sterilized with 2 % (v/v) sodium hypochlorite solution
dextrose agar (PDA) (Acumedia). PDA was autoclaved at 103 KPa (Panebianco et al., 2014), rinsed with distilled water and air-dried at
(121 C) for 15 min. PDA plates were stored at 25 C for 7 d in room temperature. The fruits were wounded at the equatorial re-
darkness. Spores were harvested by washing the agar surface with gion (1 cm wide x 1 cm deep) and 15 mL of P. digitatum spore so-
sterile distilled water and subsequently diluted to a final concen- lution was inoculated in the wound site. Infected and control fruits
tration of 106 spore mL 1. 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
2.2. Mass spectrometry imaging (MSI) 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
MSI analysis were performed directly on the orange (Citrus cut (2 cm 2 cm) and the extractions of the secondary metabolites
sinensis) peel surface infected with P. digitatum using a Prosolia produced in the fruits were made with 5 mL ethyl acetate 100 %
DESI source Modelo Omni Spray 2D®-3201) coupled to a Thermo during 1 h in ultrasonic bath. Orange juice and PDA were extracted
Scientific QExactive® Hybrid Quadrupole-Orbitrap Mass Spec- twice with equal volumes of ethyl acetate 100 % (1:1 v/v) during 1 h
trometer. The DESI configuration used was the same set by in ultrasonic bath. All the extracts were filtered, dried under N2 and
Angolini et al. (2015) with small modifications. The methanol stored at 20 C.
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 2.7. MS/MS analysis
Scientific. The DESI-MSI data was converted into image files
using Firefly data conversion software (version 2.1.05) and Crude extracts were resuspended in MeOH HPLC and centri-
viewed using BioMap software (version 3.8.0.4) developed by fuged at 13,000 rpm for 5 min. The samples were analyzed using a
Novartis Institutes for BioMedical Research. In BioMap, color LC Agilent 1200 mass spectrometer coupled with Agilent iFunnel
scaling was adjusted to a fixed value for comparison between 6550 Q-ToF LC-MS. The electrospray ionization source operated in
the samples. 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
2.3. Large scale experiment for isolation of compounds 1, 3 and 7 pressure: 50 psi; auxiliary gas temperature: 350 C and flow of
auxiliary gas: 12 mL min 1. The analyzer time of flight (ToF) oper-
Large scale cultivation was performed by cultivating P. digitatum ated in the range m/z 50e1500. Collision Energy formula (auto MS/
in 12 L of PD (Potato-Dextrose) liquid media in sterile 1 L Erlen- MS mode): 4 V (slope)*(m/z)/100 þ 5 V (offset). Maximum 5 pre-
meyer and stored at 25 C for 10 d in darkness. After incubation, cursors per cycle were selected. 2 mL of sample were injected. Sta-
culture broth was extracted twice with equal volumes of ethyl ac- tionary phase: Thermo Scientific column Accucore C18 2.6 mm,
etate 100 % (1:1 v/v) and vacuum filtered. Solvent was removed 2.1 mm 100 mm. The mobile phase was water (A) and acetonitrile
under reduced pressure and the final extract stored at 20 C. with 0.1 % (v/v) of formic acid (B). Flow rate was 0.2 mL min 1. The
31
J.H. Costa et al. / Fungal Biology 123 (2019) 594e600
eluent profile (A:B) was: 0e10 min, gradient from 95:5 to 2:98; at m/z 519.1870, m/z 505.1712, m/z 475.1612, m/z 460.1976 and m/z
10e15 min, isocratic elution with 2:98; 15e16.2 min, gradient from 459.1664 which correspond respectively to tryptoquialanine A (1)
2:98 to 95:5; 16.2e20 min, isocratic elution with 95:5. The spectra (C27H26N4O7), tryptoquialanine B (2) (C26H24N4O7), which are
were processed with Agilent Mass Hunter Workstation Software. already described to be produced by P. digitatum (Ariza et al., 2002),
tryptoquialanone (4) (C25H22N4O6), 15-dimethyl-2-epi-fumiquina-
2.8. Molecular MS/MS network zoline A (5) (C25H25N5O4), and deoxytryptoquialanone (6)
(C25H22N4O5) that are intermediates of the tryptoquialanine
A molecular network for P. digitatum cultivated in sicilian lemon, biosynthetic pathway. Tryptoquialanines biosynthesis pathway
tangerine, orange, orange juice and PDA was created using the was well unveiled for Penicillium aethiopicum by Gao et al. (2011)
online workflow at GNPS (http://gnps.ucsd.edu). The data was and intermediates 4, 5 and 6 were also detected for P. digitatum
filtered by removing all MS/MS peaks within ±17 Da of the pre- through MSI analysis, since tryptoquialanine biosynthetic gene
cursor m/z. MS/MS spectra were window filtered by choosing only clusters of both fungi show high similarity (Zhu et al., 2017) as show
the top 6 peaks in the ± 50 Da window throughout the spectrum. in Supplementary Fig. S13. All structures (Fig. 2) have been
The data was then clustered with MS-Cluster with a parent mass confirmed through exact masses and typical fragmentation pattern
tolerance of 0.2 Da and a MS/MS fragment ion tolerance of 0.1 Da to of tryptoquialanine-like alkaloids. In MS/MS analysis, main typical
create consensus spectra. Further, consensus spectra that contained fragments were observed at [MþH]þ m/z 156.07, m/z 197.10 and m/z
less than 2 spectra were discarded. A network was then created 213.10.
where edges were filtered to have a cosine score above 0.7 and In Fig. 1 it is observed that the alkaloids signals are only present
more than 6 matched peaks. Further edges between two nodes on the infected surface, indicating that these compounds are pro-
were kept in the network only if each of the nodes appeared in each duced during the mycelial growth of P. digitatum. It is also noticed
other's respective top 10 most similar nodes. The spectra in the an accumulation of the metabolites on the fruit surface: the in-
network were then searched against GNPS0 spectral libraries. The tensity of the signals increases in relation to the time of infection,
library spectra were filtered in the same manner as the input data. especially at 6 dpi, where there is higher concentration level of
All matches kept between network spectra and library spectra were these alkaloids in comparison to the previous stages of infection.
required to have a score above 0.7 and at least 6 matched peaks. Zhu et al. (2017) confirmed through RNA-seq analysis that the
expression of the tryptoquialanine biosynthetic genes occurs since
2.9. Evaluation of insecticidal activity of tryptoquialanine A the first day after inoculation and are in highest level after three
days, indicating a potential biological role of these compounds.
Larvae for all experiments were 2nd instars and kept under Fig. 1 also shows that the ion m/z 460.1973 relative to the inter-
controlled temperature conditions (T ¼ 27 ± 1 C), relative hu- mediate 5 presents a different spatial distribution compared to the
midity (RH ¼ 70 ± 5 %) and photoperiod of 12 h. The bioassay was other analyzed alkaloids being mainly found in the peripheral zone
performed following the criteria established by Dulmage et al. (white areas). It suggests that 5 is one of the initial intermediates in
(1990), with some modifications. Ten Ae. aegypti larvae were the biosynthetic route of tryptoquialanine A (Gao et al., 2011), and
placed in individual wells of a 6-well cell culture dish containing is in higher concentration in the youngest fungal cells.
10 mL water, 100 mL of liquid feed and the application of extract to The detection of the known tryptoquialanines A and B through
the concentration of 250 mg/mL was added to the larval water; the MSI shows that this technique can be used for searching new
negative control receiving only 1 % DMSO (Dimethylsulfoxide) at secondary metabolites. MSI analysis also revealed ions ([MþHþ] m/
the same concentrations, whose mortality did not exceed 10 %. z 433.1502 and m/z 503.1920) never before reported from
Survival was monitored for four days checking for the number of P. digitatum. These compounds were isolated and investigated by
live and dead larvae of each well. Three independent biological NMR spectroscopy, since they are not part of the described tryp-
replicates (with three technical replicates each) where performed toquialanine biosynthetic pathway.
for each experiment.
3.2. Isolation and characterization of secondary metabolites 1, 3
3. Results and discussion and 7
3.1. MSI analysis of green mold on orange fruits surface Secondary metabolites are small organic molecules produced by
various microorganisms and some of them have been shown to
MSI has become an important tool for mycologists since it is able contribute to the pathogenicity of several pathogenic fungi (Scharf
to image thousands of molecules, including metabolites, proteins, et al., 2014). The discovery of new secondary metabolites produced
lipids and peptides, providing a visualization of the spatial distri- by P. digitatum and the understanding of their biological role pro-
bution of secondary metabolites on fungal cultures (Sica et al., vide more information about the infection process and how this
2014; Buchberger et al., 2018). A few P. digitatum secondary me- fungus may survive the defense mechanisms of fruit. P. digitatum
tabolites are described, however, their biological roles in infection extract was obtained from a scaled up experiment and metabolites
have not yet been fully unveiled (Zhu et al., 2017) as the pathoge- of interest were isolated by preparative HPLC. Compounds 3 (m/z
nicity mechanisms and host specificity (Marcet-Houben et al., 503.1920) and 7 (m/z 433.1503) were detected first by our MSI
2012). MSI technique was applied here for the first time to inves- technique and were isolated for structural characterization.
tigate the metabolites produced by P. digitatum on citrus fruits Compound 3, previously isolated by Gao et al. (2011), is not part
surface during infection process and how they are distributed on of the normal biosynthetic pathway of tryptoquialanine A, since it
the infected fruits, giving initial insights into the role of secondary is a final product formed after the deletion of the TaqE enzyme
metabolites in the infection mechanisms of P. digitatum in the host. responsible for catalyzing a N-hydroxylation and is not yet
DESI-MSI was applied on oranges infected with P. digitatum at described as a natural product. Gao et al. (2011) suggested that the
different stages of the infection process (from 4 to 7 dpi) and an remaining tailoring steps in the tqa pathway can function in the
orange control (Fig. 1). It was possible to detect, in all infection absence of N-hydroxylation, leading to 3. So, the detection of 3 in
stages, the tryptoquialanines A and B and their biosynthetic in- the wild-type P. digitatum through MSI analysis warranted more
termediates. Fig. 1 shows the MSI signals obtained for ions [MþH]þ 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 presented three singlets and one doublet of methyl groups, one
by NMR spectroscopy, confirming that it was a new secondary methylene group, three methine groups at dH 5.30 (H-2), 5.40 (H-
metabolite produced by P. digitatum. Supplementary Table S1 12) and 6.25 (H-27), and signals corresponding to nine aromatic
shows NMR data obtained for 3. The NMR signals for 3 are protons between dH 7.30 and 8.40. Signals characteristic of an imine
similar to those for 1 (Ariza et al., 2002), with exception of the at dc 153.2 (C-26) and of one amide group at dc 161.7 (C-18) sug-
presence of an NH signal at 3.05 ppm and the absence of an NOH gested that the other aromatic ring was derived from a part bound
signal (dH 7.95) (Supplementary Fig. S12). 1H NMR spectrum 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.
33
J.H. Costa et al. / Fungal Biology 123 (2019) 594e600
structure and the rest of the molecule structure can inferred by limited to the analysis of the fruits surface, so that internal me-
HMBC correlations from dc 153.2 (C-26) to dH 6.25 (H-27) and from tabolites remain undetected (Sica et al., 2014). Other complemen-
dc 161.7 (C-18) to dH 5.40 (H-12). Long-range HMBC correlations tary approaches are needed to further characterize the metabolic
were also observed from dc 133.9 (C-4) to dH 3.05 (NeH) and from a profile during infection. Therefore, molecular networking was used
lactone carbonyl group at dc 162.5 (C-11) to the singlet at dH 5.30 (H- here, as a complement to DESI-IMS, to annotate secondary me-
2) (Ariza et al., 2002), allowing elucidation of 3. tabolites produced during P. digitatum's infection.
Through 1H and 13C NMR data (Supplementary Table S2) for 7, it Another alternative to identify unknown compounds, involves
was possible to conclude that compound 7 is tryptoquivaline L, metabolite fragmentation patterns acquired through tandem MS
alkaloid first isolated from Aspergillus fumigatus and well charac- analysis, that can be matched to those presented in databases such
terized in the literature (Yamazaki et al., 1979) (Supplementary as GNPS, PubChem and others useful databases for dereplication
Fig. S16 ans S17). Tryptoquivaline L has never have been reported, (Covington et al., 2017; Smith et al., 2005). Nowadays, the largest
until now, as a P. digitatum metabolite in the literature. natural product public database with MS/MS spectra is GNPS:
Compound 1 was also isolated and its structure was confirmed Global Natural Products Social Molecular Networking, containing
through 1H and 13C NMR data (Supplementary Figs. S19 and S20). The more than 140.000 natural products (Covington et al., 2017;
NMR signals obtained for 1 were in agreement with the data already Gaude ^ncio and Pereira, 2015).
reported in the literature for this compound (Ariza et al., 2002). 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
3.3. Evaluation of insecticidal activity of tryptoquialanine A
(Ghooshkhaneh et al., 2018). The metabolic profiles of P. digitatum
in distinct citrus fruits (sicilian lemon, tangerine and orange), or-
Zhu et al. (2017) evaluated the involvement of tryptoquialanines
ange juice and PDA were analyzed in the GNPS database through
produced P. digitatum in the citrus infection process. The authors
tandem MS (MS/MS) aimed at finding new metabolites.
deleted the tqaA gene (Non-ribosomal peptide synthetase)
After generating the molecular networks, the node connectivity
responsible for tryptoquialanine A production. Phenotype assays
was visualized (Supplementary Fig. S22). Each node represents one
suggested no significant variation between the non-TQA mutant
MS/MS spectrum and is labeled with the parent (precursor) mass.
and the wild-type strain in fungal growth, concluding that tryp-
Since fragmentation spectra generally reflect the chemical struc-
toquialanine A was not involved in the pathogenicity of P. digitatum
tures of the fragmented ions, it becomes possible to represent the
and does not act as a virulence factor, implying a different biological
produced metabolites in clusters of similar structures through the
role for the indole alkaloids (Zhu et al., 2017).
GNPS platform (Nguyen et al., 2013) (as observed in Fig. 4). For
Based on structure similarity, tryptoquialanine A (1) is consid-
P. digitatum extracts the network contained 31 different clusters.
ered a tremorgenic mycotoxin (Ariza et al., 2002) acting on the
The visualization of cluster B showed that the ions [MþH]þ m/z
central nervous system of vertebrate animals, causing symptoms
519.1871 and 505.1712, clustered together in the molecular network
such as decreased activity and immobility (Gao et al., 2011). Since 1
(Fig. 4), are tryptoquialanine A 1 and tryptoquialanine B 2 respec-
is not related to the P. digitatum pathogenicity, Zhu et al. (2017)
tively, major secondary metabolites for P. digitatum (Ariza et al.,
suggests that it may have an insecticidal biological function
2002), and observed in all growth media. Interestingly, 1 and 2
within its micro ecosystem.
were observed in vivo in all citrus fruits, suggesting an important
Compound 1 was isolated and applied in insecticidal assays to
biological role, which was confirmed here through the insecticidal
evaluate the hypothesis suggested by Zhu et al. (2017). In 24 h after
biossays.
the exposure to 1, the mortality rate of Ae. Aegypti larvae was 37 %.
The tryptoquialanine intermediates detected by MSI analysis
The larvae mortality gradually increased during the days and
were also observed in P. digitatum extracts. 5 (m/z 460.1976) is one
reached 81 % on the fourth day (Fig. 3). This result indicates a very
of the precursors of tryptoquialanine pathway and was observed in
pronounced insecticidal activity of 1 and suggests that 1 may be
orange and PDA media. Similarly, 6 (m/z 459.1664) and 3 (m/z
important as a biocontrol against insects during orange decay.
503.1920) were also observed, being the last one found in all
growth media and 6 was only produced in sicilian lemon and PDA
3.4. Comparing alkaloids in different citrus: molecular network media. Finally, 4 (m/z 475.1612), another intermediate involved in
the tryptoquialanine biosynthesis, was only detected in PDA under
Although DESI-MSI is a great tool to discover new secondary the cultivation conditions tested.
metabolites involved in biological interactions, the technique is 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
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
siamenisis KUFC 6349 (Buttachon et al., 2012). This compound was
samples were evaluated by unpaired t test (**, p < 0.05, ***p < 0.05, ****p < 0.0001), also isolated from marine sponge associated fungus Neosartorya
which had significant effects on mortality. paulistensis KUFC 7897 (Gomes et al., 2014) and the marine-derived
34
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 study of the metabolic profile produced by P. digitatum in different
well as Neosartorya takakii KUFC 7898 (Zin et al., 2015). Trypto- citrus hosts, confirming the production of known and new alkaloids
quivaline Q (8) has been previously isolated from a marine-derived in the mycelium of P. digitatum. Therefore, MSI and Molecular
fungus Neosartorya sp. HN-M-3 (Sun et al., 2012). Regarding the Networking taken together, provide a solid insight into different
fumiquinazolines A (9) and C (10), Numata et al., (1992) isolated nature interactions and pathogen infection process. Through these
both molecules from the mycelium of A. fumigatus. These com- complementary approaches it was possible to provide the first in-
pounds exhibited moderate cytotoxicity against the cultured P-388 sights about the indole alkaloids involved in the P. digitatum-citrus
lymphocytic leukemia cells (Numata et al., 1992). Fumiquinazolines interaction and initial thoughts of the possible biological role
and tryptoquialanines present similar enzymes in their biosyn- associated to them. In addition, the knowledge concerning the
thetic pathway and also have Fumiquinazoline F as an intermediate potential biological role of tryptoquialanines can serve as a target to
in common (Gao et al., 2011; Ames et al., 2011). Sequence analysis the development of new biological insecticides to be potentially
of TqaA enzyme from P. aethiopicum tryptoquialanine pathway used in agriculture. In conclusion, the secondary metabolism of
revealed high homology with Af12080 enzyme from A. fumigatus P. digitatum and the roles of natural products in the infection are
fumiquinazoline pathway. TqAa enzyme, in the tryptoquialanine still an open field for new research opportunities.
pathway, catalyzes the biosynthesis of the Fumiquinazoline F, in-
termediate that can be used in both pathways (Gao et al., 2011). Acknowledgements
Therefore, is not surprising to detect fumiquinazolines in
P. digitatum's extract when we analyze both biosynthetic pathways. This study was financed in part by the Coordenaça ~o de Aper-
Until now, 8, 9 and 10 have never been reported as P. digitatum feiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance
secondary metabolites. Code 001, Fundaç~ao de Amparo a Pesquisa no Estado de Sa ~o Paulo
GNPS molecular networking allowed consistent identification of [grant number FAPESP 2017/24462-4 and 2018/03670-0] and Pro -
metabolites present not only on the surface of the fruit but also in Reitoria de Pesquisa Unicamp (PRP/FAEPEX). MCFP is recipient of a
the different P. digitatum extracts. The detection of ions through PhD Fellowship from CAPES (grant number 8887.137194/2017-00).
MSI and GNPS analyses confirm the production of known and new JASN is recipient of a FAPESP Young Investigator Award (grant
indole alkaloids in the mycelium of P. digitatum. number 2013/11343-6).
Despite great economic interest related to P. digitatum, studies Supplementary data to this article can be found online at
concerning this pathogen-host system have mainly focused on https://doi.org/10.1016/j.funbio.2019.03.002.
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. References
In this context, MSI has emerged as a powerful tool to gain more
Ames, B.D., Haynes, S.W., Gao, X., Evans, B.S., Kelleher, N.L., Tang, Y., Walsh, C.T.,
insights concerning the natural products produced in P. digitatum 2011. Complexity generation in fungal peptidyl alkaloid biosynthesis: oxidation
infection that may have an important biological role. The MSI of fumiquinazoline a to the heptacyclic hemiaminal fumiquinazoline C by the
analysis enabled monitoring and understanding the indole alkaloid flavoenzyme Af12070 from Aspergillus fumigatus. Biochemistry 50 (4), 8756-
8787-69. http://doi.org/10.1021/bi201302w.
production by P. digitatum during infection process in citrus. Angolini, C.F.F., Vendramini, P.H., Araújo, F.D.S., Araújo, W.L., Augusti, R.,
Complementarily to MSI, GNPS molecular networking allowed the Eberlin, M.N., De Oliveira, L.G., 2015. Direct protocol for ambient mass
35
J.H. Costa et al. / Fungal Biology 123 (2019) 594e600
spectrometry imaging on agar culture. Anal. Chem. 87, 6925e6930. https://doi. Penicillium digitatum, the main postharvest pathogen of citrus. BMC Genomics
org/10.1021/acs.analchem.5b01538. 13, 646. https://doi.org/10.1186/1471-2164-13-646.
Ariza, M.R., Larsen, T.O., Petersen, B.O., Duus, J.Ø., Barrero, A.F., 2002. Penicillium Nguyen, D.D., Wu, C.H., Moree, W.J., Lamsa, A., Medema, M.H., Zhao, X.,
digitatum metabolites on synthetic media and citrus fruits. J. Agric. Food Chem. Gavilan, R.G., Aparicio, M., Atencio, L., Jackson, C., Ballesteros, J., Sanchez, J.,
50, 6361e6365. https://doi.org/10.1021/jf020398d. Watrous, J.D., Phelan, V.V., van de Wien, C., Kersten, R.D., Mehnaz, S., De Mot, R.,
Barmore, C.R., Brown, G.E., 1981. Polygalacturonase from citrus fruit infected with Shank, E.A., Charusanti, P., Nagarajan, H., Duggan, B.M., Moore, B.S., Bandeira, N.,
Penicillium italicum. Phytopathology 71, 328e331. Palsson, B.O., Pogliano, K., Gutie rrez, M., Dorrestein, P.C., 2013. MS/MS
Buchberger, A.R., DeLaney, K., Johnson, K., Li, L., 2018. Mass spectrometry imaging: a networking guided analysis of molecule and gene clusters families. Proc. Natl.
review of emerging advancements and future insights. Anal. Chem. 90, Acad. Sci. U. S. A. 110, E2611eE2620. https://doi.org/10.1073/pnas.1303471110.
240e265. http://doi.org/10.1021/acs.analchem.7b04733. Numata, A., Takahashi, C., Matsushita, T., Miyamoto, M., Kawai, K., Usami, Y.,
Buttachon, S., Chandrapatya, A., Manoch, L., Silva, A., Gales, L., Bruye re, C., Kijjoa, A., Matsumura, E., Inoue, M., Ohishi, H., Shingu, T., 1992. Fumiquinazolines, novel
2012. Sartorymensin, a new indole alkaloid, and new analogues of tryptoqui- metabolites of a fungus isolates from a saltfish. Tetrahedron Lett. 33 (12),
valine and fiscalins produced by Neosartorya siamensis (KUFC 6349). Tetrahe- 1621e1624. https://doi.org/10.1016/S0040-4039(00)91690-3.
dron 68, 3253e3262. Panebianco, S., Vitale, A., Platania, C., Restuccia, C., Polizzi, G., Cirvilleri, G., 2014.
Covington, B.C., McLean, J.A., Bachmann, B.O., 2017. Comparative mass Postharvest efficacy of resistance inducers for the control of green mold on
spectrometry-based metabolomics strategies for the investigation of microbial important Sicilian citrus varieties. J. Plant Dis. Prot. 121, 177e183. https://doi.
secondary metabolites. Nat. Prod. Rep. 34, 6e24. https://doi.org/10.1039/ org/10.1007/BF03356507.
C6NP00048G. Perez, M.F., Ibarreche, J.P., Isas, A.S., Sepulveda, M., Ramallo, J., Dib, J.R., 2017.
Dulmage, H.T., Yousten, A.A., Singer, S., Lacey, L.A., 1990. Guidelines for Production Antagonistic yeasts for the biological control of Penicillium digitatum on
of Bacillus Thuringiensis H-14 and Bacillus Sphaericus. UNDP/World Bank/WHO, lemons stored under export conditions. Biol. Control 115, 135e140. https://doi.
Steering Committee to Biological Control of Vetores, Geneva, p. 59. org/10.1016/j.biocontrol.2017.10.006.
Eamvijarn, A., Gomes, N.M., Dethoup, T., Buaruang, J., Manoch, L., Silva, A., Pedro, M., Prusky, D., McEvoy, J.L., Saftner, R., Conway, W.S., Jones, R., 2004. Relationship be-
Marini, I., Roussis, V., Kijjoa, A., 2013. Bioactive meroditerpenes and indole al- tween host acidification and virulence of Penicillium spp. on apple and citrus
kaloids from the soil fungus Neosartorya fischeri (KUFC 6344), and the marine- fruit. Phytopathology 94, 44e51. http://doi.org/10.1094/PHYTO.2004.94.1.44.
derived fungi Neosartorya laciniosa (KUFC 7896) and Neosartorya tsunodae Scharf, D.H., Heinekamp, T., Brakhage, A.A., 2014. Human and plant fungal patho-
(KUFC 9213). Tetrahedron 69, 8583e8591. gens: the role of secondary metabolites. PLoS Pathog. 10 (1), e1003859. https://
Gao, X., Chooi, Y.H., Ames, B.D., Wang, P., Walsh, C.T., Tang, Y., 2011. Fungal indole doi.org/10.1371/journal.ppat.1003859.
alkaloid biosynthesis: genetic and biochemical investigation of the trypto- Sica, V.P., Raja, H.A., El-Elimat, T., Oberlies, N.H., 2014. Mass spectrometry imaging of
quialanine pathway in Penicillium aethiopicum. J. Am. Chem. Soc. 133 (8), secondary metabolites directly on fungal cultures. RSC Adv. 4, 63221e63227.
2729e2741. https://doi.org/10.1021/ja1101085. https://doi.org/10.10139/c4ra11564c.
Gaude ^ncio, S.P., Pereira, F., 2015. Dereplication: racing to speed up the natural Smith, C.A., O'Maille, G., Want, E.J., Qin, C., Trauger, S.A., Brandon, T.R., Siuzdak, G.,
products discovery process. Nat. Prod. Rep. 32, 779e810. 2005. METLIN: a metabolite mass spectral database. Ther. Drug Monit. 27,
Gomes, N.M., Bessa, L.J., Buttachon, S., Costa, P.M., Buaruang, J., Dethoup, T., Silva, A., 747e751.
Kijjoa, A., 2014. Antibacterial and antibiofilm activities of tryptoquivalines and Sun, F., Chen, G., Bai, J., Pei, Y., 2012. Two new alkaloids from a marine-derived
meroditerpenes isolated from the marine-derived fungi Neosartorya paulis- fungus Neosartorya sp. HN-M-3. J. Asian Nat. Prod. Res. 14 (12), 1109e1115.
tensis, N. laciniosa, N. tsunodae, and the soil fungi N. fischeri and N. siamensis. Yamazaki, M., Fujimoto, H., Okuyama, E., 1978. Structure determination of six fungal
Mar. Drugs 12, 822e839. metabolites, tryptoquivaline E, F, G, H, I and J from Aspergillus fumigatus. Chem.
Ghooshkhaneh, N.G., Golzarian, M.R., Mamarabadi, M., 2018. Detection and classifi- Pharm. Bull. 26, 111e117.
cation of citrus green mold caused by Penicillium digitatum using multispectral Yamazaki, M., Okuyama, E., Maebayashi, Y., 1979. Isolation of some new
imaging. J. Sci. Food Agric. 98, 3542e3550. https://doi.org/10.1002/jsfa.8865. tryptoquivaline-related metabolites from Aspergillus fumigatus. Chem. Pharm.
Lei, Z., Huhman, D., Sumner, L.W., 2011. Mass spectrometry strategies in metab- Bull. 27, 1611e1617.
olomics. J. Biol. Chem. jbc-R111 https://doi.org/10.1074/jbc.R111.238691. Zhang, T., Sun, X., Xu, Q., Candelas, L.G., Li, H., 2013. The pH signaling transcription
Lopez-Pe rez, M., Ballester, A., Gonz alez-Candelas, L., 2015. Identification and func- factor PacC is required for full virulence in Penicillium digitatum. Appl. Micro-
tional analysis of Penicillium digitatum genes putatively involved in virulence biol. Biotechnol. 97, 9087e9098. https://doi.org./10.1007/s00253-013-5129-x.
towards citrus fruit. Mol. Plant Pathol. 16, 262e275. https://doi.org/10.1111/ Zhu, C., Sheng, D., Wu, X., Wang, M., Hu, X., Li, H., Yu, D., 2017. Identification of
mpp.12179. secondary metabolite biosynthetic gene clusters associated with the infection
Macarisin, D., Cohen, L., Eick, A., Rafael, G., Belausov, E., Wisniewski, M., Droby, S., of citrus fruit by Penicillium digitatum. Postharvest Biol. Technol. 134, 17e21.
2007. Penicillium digitatum suppresses production of hydrogen peroxide in https://doi.org/10.1016/j.postharvbio.2017.07.011.
host tissue during infection of citrus fruit. Phytopathology 97, 1491e1500. Zin, W.W.M., Buttachon, S., Buaruang, J., Gales, L., Pereira, J.A., Pinto, M.M.M.,
https://doi.org/10.1094/PHYTO-97-11-1491. Silva, A.M.S., Kijjoa, A., 2015. A new meroditerpene and a new tryptoquivaline
Marcet-Houben, M., Ballester, A.R., La Fuente, B., Harries, E., Marcos, J.F., Gonza lez- analog from the algicolous fungus Neosartorya takakii KUFC 7898. Mar. Drugs
Candelas, L., Gabaldo n, T., 2012. Genome sequence of the necrotrophic fungus 13 (6), 3776e3790. https://doi.org/10.3390/md13063776.
36
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*
Fig. S1. Oranges infected with P. digitatum used for MSI analysis.
Fig. S3. Photograph of DESI source on sample of orange peel infected with P. digitatum.
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
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. 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
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.
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. 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
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
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
g Instituto de Microbiologia Paulo de Góes (IMPG), Universidade Federal do Rio de Janeiro, Rio de Janeiro, Rio de Janeiro, Brazil
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
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).
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
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
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.
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
(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
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.
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
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.
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
REFERENCES
1. Jr-Mattos D, Carlos EF. 2018. The role of the International Society of Citri- DsRed and GFP to visualize citrus host colonization. J Microbiol Methods
culture on the world citrus industry. Citrus Res Technol 38:228–232. 144:134–144. https://doi.org/10.1016/j.mimet.2017.11.019.
https://doi.org/10.4322/crt.ICC171. 15. Scharf DH, Heinekamp T, Brakhage AA. 2014. Human and plant fungal
2. Citrus: World Markets and Trade. 2020. United States Department of Agri- pathogens: the role of secondary metabolites. PLoS Pathog 10:e1003859.
culture. Foreign Agricultural Service. https://downloads.usda.library https://doi.org/10.1371/journal.ppat.1003859.
.cornell.edu/usda-esmis/files/w66343603/00000g55g/kp78h0193/citrus 16. Ariza MR, Larsen TO, Peterson BO, Duus JO, Barrero AF. 2002. Penicillium
.pdf. Accessed 14 April 2020. digitatum metabolites on synthetic media and citrus fruits. J Agric Food
3. Blauth de Lima F, Félix C, Osório N, Alves A, Vitorino R, Domingues P, Chem 50:6361–6365. https://doi.org/10.1021/jf020398d.
Correia A, Ribeiro RTS, Esteves AC. 2016. Secretome analysis of Tricho- 17. Costa JH, Bazioli JM, Araújo EV, Vendramini PH, Porto MCF, Eberlin MN,
derma atroviride T17 biocontrol of Guignardia citricarpa. Biol Control Souza-Neto JA, Fill TP. 2019. Monitoring indole alkaloid production by Peni-
99:38–86. https://doi.org/10.1016/j.biocontrol.2016.04.009. cillium digitatum during infection process in citrus by mass spectrometry
4. Costa JH, Bazioli JM, Pontes JGM, Fill TP. 2019. Penicillium digitatum infec- imaging and molecular networking. Fungal Biol 123:594–600. https://doi
tion mechanisms in citrus: what do we know so far? Fungal Biol .org/10.1016/j.funbio.2019.03.002.
123:584–593. https://doi.org/10.1016/j.funbio.2019.05.004. 18. Costa JH, Wassano CI, Angolini CFF, Scherlach K, Hertweck C, Fill TP. 2019.
5. Qian X, Yang Q, Zhang Q, Abdelhai MH, Dhanasekaran S, Serwah BNA, Antifungal potential of secondary metabolites involved in the interaction
Gu N, Zhang H. 2019. Elucidation of the initial growth process and the between citrus pathogens. Sci Rep 9:18647. https://doi.org/10.1038/
infection mechanism of Penicillium digitatum on postharvest citrus s41598-019-55204-9.
(Citrus reticulata blanco). Microorganisms 7:485. https://doi.org/10 19. Souza JAM, Baltazar LM, Carregal VM, Gouveia-Eufrasio L, Oliveira AG,
.3390/microorganisms7110485. Dias WG, Rocha MC, Miranda KR, Malavazi I, Santos DA, Frézard FJG,
6. Chen J, Shen Y, Chen C, Wan C. 2019. Inhibition of key citrus postharvest Souza DG, Teixeira MM, Soriani FM. 2019. Characterization of Aspergillus
fungal strains by plant extracts in vitro and in vivo: a review. Plants 8:26. fumigatus extracellular vesicles and their effects on macrophages and
https://doi.org/10.3390/plants8020026. neutrophils functions. Front Microbiol 10:2008. https://doi.org/10.3389/
7. Zhang S, Zheng Q, Xu B, Liu J. 2019. Identification of the fungal pathogens fmicb.2019.02008.
of postharvest disease on peach fruits and the control mechanisms of Bacil- 20. Herkert PF, Amatuzzi RF, Alves LR, Rodrigues ML. 2019. Extracellular
lus subtilis JK-14. Toxins 11:322. https://doi.org/10.3390/toxins11060322. vesicles as vehicles for the delivery of biologically active fungal mole-
8. Spadaro D, Droby S. 2016. Development of biocontrol products for post- cules. Curr Protein Pept Sci 20:1027–1036. https://doi.org/10.2174/
harvest diseases of fruit: the importance of elucidating the mechanisms 1389203720666190529124055.
of action of yeast antagonists. Trends Food Sci Technol 47:39–49. https:// 21. Silva VKA, Rodrigues ML, May RC. 2019. Deciphering fungal extracellular
doi.org/10.1016/j.tifs.2015.11.003. vesicles: from cell biology to pathogenesis. Curr Clin Micro Rep 6:89–97.
9. Sanzani SM, Reverberi M, Geisen R. 2016. Mycotoxins in harvested fruits https://doi.org/10.1007/s40588-019-00128-1.
and vegetables: insights in producing fungi, biological role, conducive 22. Rybak K, Robatzek S. 2019. Functions of extracellular vesicles in immunity
conditions, and tools to manage postharvest contamination. Postharvest and virulence. Plant Physiol 179:1236–1247. https://doi.org/10.1104/pp
Biol Tec 122:95–105. https://doi.org/10.1016/j.postharvbio.2016.07.003. .18.01557.
10. Nicosia MGLD, Pangallo S, Raphael G, Romeo FV, Strano MC, Rapisarda P, 23. Regente M, Pinedo M, Clemente HS, Balliau T, Jamet E, Canal L. 2017.
Droby S, Schena L. 2016. Control of postharvest fungal rots on citrus fruit Plant extracellular vesicles are incorporated by a fungal pathogen and in-
and sweet cherries using a pomegranate peel extract. Postharvest Biol hibit its growth. J Exp Bot 68:5485–5496. https://doi.org/10.1093/jxb/
Tec 114:54–61. https://doi.org/10.1016/j.postharvbio.2015.11.012. erx355.
11. Dukare AS, Paul S, Nambi VE, Gupta RK, Singh R, Sharma K, Vishwakarma 24. Bielska E, Sisquella MA, Aldeieg M, Birch C, O’Donoghue EJ, May RC. 2018.
RK. 2019. Exploitation of microbial antagonists for the control of post- Pathogen-derived extracellular vesicles mediate virulence in the fatal
harvest diseases of fruits: a review. Crit Rev Food Sci Nutr 59:1498–1513. human pathogen Cryptococcus gattii. Nat Commun 9:1556. https://doi
https://doi.org/10.1080/10408398.2017.1417235. .org/10.1038/s41467-018-03991-6.
12. Bazioli JM, Belinato JR, Costa JH, Akiyama DY, Pontes JGM, Kupper KC, 25. Bleackley MR, Samuel M, Garcia-Ceron D, McKenna JA, Lowe RGT, Pathan
Augusto F, Carvalho JE, Fill TP. 2019. Biological control of citrus postharvest M, Zhao K, Ang C-S, Mathivanan S, Anderson MA. 2019. Extracellular
phytopathogens. Toxins 11:460. https://doi.org/10.3390/toxins11080460. vesicles from the cotton pathogen Fusarium oxysporum f. sp. vasinfectum
13. Ramón-Carbonell M, Sánchez-Torres P. 2017. The transcription factor induce a phytotoxic response in plants. Front Plant Sci 10:1610. https://
PdSte12 contributes to Penicillium digitatum virulence during citrus doi.org/10.3389/fpls.2019.01610.
fruit infection. Postharvest Biol Tec 125:129–139. https://doi.org/10 26. Reis FCG, Borges BS, Jozefowicz LJ, Sena BAG, Garcia AWA, Medeiros LC,
.1016/j.postharvbio.2016.11.012. Martins ST, Honorato L, Schrank A, Vainstein MH, Kmetzsch L, Nimrichter
14. Vu TX, Ngo TT, Mai LTD, Bui T, Le DH, Bui HTV, Nguyen HQ, Ngo BX, Tran L, Alves LR, Staats CC, Rodrigues ML. 2019. A novel protocol for the isola-
V. 2018. A highly efficient Agrobacterium tumefaciens-mediated transfor- tion of fungal extracellular vesicles reveals the participation of a putative
mation system for the postharvest pathogen Penicillium digitatum using scramblase in polysaccharide export and capsule construction in
Cryptococcus gattii. mSphere 4:e00080-19. https://doi.org/10.1128/mSphere vera gel alone or with the addition of thymol. Int J Food Microbiol
.00080-19. 151:241–246. https://doi.org/10.1016/j.ijfoodmicro.2011.09.009.
27. Gao X, Chooi YH, Ames BD, Wang P, Walsh CT, Tang Y. 2011. Fungal indole 40. Louw JP, Korsten L. 2016. Postharvest decay of nectarine and plum
alkaloid biosynthesis: genetic and biochemical investigation of the tryp- caused by Penicillium spp. Eur J Plant Pathol 146:779–791. https://doi.org/
toquialanine pathway in Penicillium aethiopicum. J Am Chem Soc 10.1007/s10658-016-0956-0.
133:2729–2741. https://doi.org/10.1021/ja1101085. 41. Louw JP, Korsten L. 2019. Impact of ripeness on the infection and coloni-
28. Bertinetti BV, Peña NI, Cabrera GM. 2009. An antifungal tetrapeptide from sation of Penicillium digitatum and P expansum on plum. Postharvest Biol
the culture of Penicillium canescens. Chem Biodivers 6:1178–1184. https:// Technol 149:148–158. https://doi.org/10.1016/j.postharvbio.2018.11.024.
doi.org/10.1002/cbdv.200800336. 42. Niedz RP. 2008. In vitro germination of citrus seed. Proc Fla State Hort Soc
29. Hammerl R, Frank O, Schmittnägel T, Ehrmann MA, Hofmann T. 2019. 121:148–151.
Functional metabolome analysis of Penicillium roqueforti by means of dif- 43. Habermann E, Pereira VDC, Imatomi M, Pontes FC, Gualtieri CJ. 2017. In
ferential off-line LC-NMR. J Agric Food Chem 67:5135–5146. https://doi vitro herbicide activity of crude and fractionated leaf extracts of Blepharo-
.org/10.1021/acs.jafc.9b00388. calyx salicifolius (Myrtaceae). Braz J Bot 40:33–40. https://doi.org/10.1007/
30. Zhu C, Sheng D, Wu X, Wang M, Hu X, Li H, Yu D. 2017. Identification of s40415-016-0317-4.
secondary metabolite biosynthetic gene clusters associated with the 44. Abdelgaleil SAM, Saad MMG, Ariefta NR, Shiono Y. 2020. Antimicrobial
infection of citrus fruit by Penicillium digitatum. Postharvest Biol Tec and phytotoxic activities of secondary metabolites from Haplophyllum
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#
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 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
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
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
⁎
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 a battlefield in which the pathogen attacks and the plant activates de-
fense mechanisms [7]. For citrus fruit, it is well known the production
Citrus have the largest global production compared to other fruit of phenylpropanoids as part of the defense mechanism [8,9,10]. The
genus and the forecast for citrus global production in 2019/2020 is citrus fruit peel is rich in flavanones, flavones and polymethoxylated
about 95 million metric tons (tons), including oranges, lemons, tan- flavones, such as naringin, hesperidin, diosmin, tangeretin and nobi-
gerines and grapefruit [1]. Therefore, citriculture has an important letin [8,11], that have been associated to defense against phytopatho-
economic impact worldwide [2], especially for the largest producers gens, especially, P. digitatum [8,9,11]. Arcas et al. (2000) [10] reported
such as Brazil, China and United States [1]. that fruit exposed to 2 h of UV light had the levels of naringin and
Since citrus are rich in water and nutrients, these fruits are predis- tangeretin increased by 7 and 55% respectively, and the growth of P.
posed to be infected by phytopathogens during postharvest period [2] digitatum in these fruits was reduced up to 45% when compared to non-
and the fungal diseases are the main cause of citrus losses [3]. The main irradiated fruit. In another study, Kim et al. (2011) [9] showed that the
fungi pathogens of citrus are Penicillium digitatum, Penicilium italicum flavonoid level in fruit infected with P. digitatum decrease in the be-
and Geotrichum citri-aurantii, which cause the green mold, blue mold ginning of the infection, then gradually increase in the intermediate
and sour rot diseases, respectively [2,4]. These diseases account for up stage, before decreasing in the final stage. The initially decrease is due
to 80% of the postharvest losses in citrus, leading to huge economic to the known hydrolyzing action of P. digitatum that transform the
losses [3]. glycosylated flavanones to the corresponding aglycones [8,9,10,12],
Nowadays, the prevent of postharvest diseases caused by fungi in while the increase in the intermediate infection stage, are associated to
citrus is through the use of chemical fungicides, such as imizalil, the defense mechanisms of the plant [9].
leading to concerns about fungi resistance and environmental and Literature only describes the hydrolyzing ability of P. digitatum,
health risks [4,5]. In order to develop alternative and safer strategies to however other important citrus pathogens, such as P. italicum, have not
protect citrus, it is necessary to elucidate the chemical interaction and been studied concerning such enzymes and mechanisms. Furthermore,
virulence strategies that occur during the host-pathogen interaction [6]. other P. digitatum’s enzymatic activities on the citrus flavonoids were
The interaction between plant and pathogen during the infection is 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
and the flavonoids are important to understand the infection process, incubated in individual sterile 500 mL beakers (darkness at 25 °C)
the objective of this study was to compare the hydrolysis activity of the during 3 or 5 d. Assays were performed in triplicate.
main citrus pathogens (P. digitatum, P. citrinum, P. italicum and After the incubation period (3 or 5 d), the oranges peels (4 × 4 cm)
Geotrichum citri-aurantii) and one non-citrus pathogen (P. expansum) were cut on the infected area nearest the inoculation wound.
through in vitro biocatalysis using the main citrus flavonoids (hesper- Extractions were performed following methodology described by
idin and naringin) as substrates. In addition, in vivo assays and mole- Smedsgaard (1997) [15] with few modifications. 0.6 g of orange peel
cular networking was employed to investigate other biotransformations were extracted with 5 mL of methanol during 1 h in ultrasonic bath.
in citrus flavonoids during host-pathogen interaction. 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. Material and methods
2.5. Mass spectrometry analysis
2.1. Commercial reagents
For in vitro assays, HPLC-MS analysis were performed using a Waters
Commercially hesperidin, hesperitin, and naringin were purchased ACQUITY UPLC system coupled to a Waters Micromass Quattro Micro
from Sigma-Aldrich (St Luis, MO, USA). TM API, with electrospray ionization source and triple quadrupole mass
analyzer. Analyses were carried out in the positive mode with m/z
2.2. Fungal strains and culture conditions 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
P. citrinum (PC) and Geotrichum citri-aurantii (GC) strains were ob- 200 °C. Stationary phase was Thermo Scientific column Accucore C18
tained from Sylvio Moreira Citrus Center (Cordeirópolis, SP, Brazil). P. (2.1 mm × 100 mm × 2.6 µm). Mobile phase was 0.1% formic acid (A)
digitatum (PD) and P. italicum (PI) strains are deposited in the Spanish and acetonitrile (B). Eluent profile (A/B): 95/5 up to 2/98 within
Type Culture Collection (CECT) (accession code CECT20796 and 10 min, held for 5 min, up to 95/5 within 1.2 min and held for 3.8 min.
CECT20909, respectively). P. expansum (PE) are deposited in the The total run time was 20 min for each run and the flow rate,
Tropical Culture Collection from André Tosello Collection (Campinas, 0.2 mL min−1. Injection volume was 5 µL.
SP, Brazil) (accession code CCT 4680). For in vivo assays, high resolution mass spectrometry analyses
Fungi were cultured on potato dextrose agar (7 d at 25 °C). Spore (HPLC-HRMS) were performed in a Waters Acquity UPLC H-class,
suspensions were prepared using sterile distilled water and were ad- Waters Xevo G2-XS QToF mass spectrometer using electrospray ioni-
justed to a final concentration of 106 spores mL−1. 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;
2.3. In vitro biotransformation of flavonoids 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
Erlenmeyer flasks (500 mL) containing 200 mL of potato dextrose mass) and 60–80 V (high mass). Stationary phase was BEH C18 column
media and 40 µL of fungi spore suspension were incubated at 25 °C and Accucore C18 (2.1 mm × 100 mm × 1.7 µm). Mobile phase was 0.1%
90 rpm during 72 h. The cells were vacuum filtered and washed with formic acid (A) and acetonitrile (B). Eluent profile (A/B): 90/10 up to
acetate buffer (pH 4.0, 0.1 M). 50/50 within 6 min, up to 2/98 within 3 min and up to 90/10 within
The biocatalysis assays were performed using the protocol described 1 min. The total run time was 10 min for each run and the flow rate,
by Costa et al. (2017) [13] with few modifications. 1 g of wet cells, 0.2 mL min−1. Injection volume was 2 µL. All the operation and spectra
20 mL of acetate buffer and 2 mg of flavonoid (diluted in 100 µL of analysis were conducted using Waters MassLynx (version 4.1).
DMSO) were added to Erlenmeyer flasks (125 mL). Negative controls
were performed using acetate buffer and flavonoids to consider possible 2.6. Antifungal assays
spontaneous hydrolysis of the flavonoids. Also, fungi controls were
performed with acetate buffer and wet cells, to consider secondary For antifungal assays, 7 mL of PDA were supplemented with 20 µL
metabolites naturally produced by each of the fungi strains. All assays of hesperidin or hesperitin solution and poured in Petri dishes.
and negative controls were performed in triplicate. The Erlenmeyer Flavonoid solutions were made with DMSO (0.3%) and final con-
flasks were incubated at 25 °C and 90 rpm during 7 d. centrations in PDA were 50, 200 and 400 mg L−1. 5 µL of spore solution
Assays and controls were monitored in the initial time and after 1, 3, (106 spores mL−1) of PD, PI, PC, PE or GC were inoculated in on the
6, 24, 48, 72, 96 and 168 h (7 d) for PD and PI, and after 6 h for PC, PE center of each agar plate. Negative controls were performed with 20 µL
and GC. Aliquots of 1 mL were saturated with NaCl, extracted twice of DMSO and the plates were incubated at 25 °C, in darkness, during 7
with ethyl acetate (0.5 mL) and centrifuged at 18,000 g. Organic layers d.
were combined, dried, resuspended with 800 µL of HPLC methanol and
analyzed by HPLC-MS. Benzophenone was used as internal standard. 2.7. Molecular network
Enzymatic conversions (%C) were calculated using chromatographic
peak area of internal standard and reagent (Eq. (1)). Aisnc and Aisr are A molecular networking for in vivo assays of P. digitatum was created
the peak area of internal standard in the negative control and in the using the online workflow at GNPS (http://gnps.ucsd.edu). The data
reaction, respectively, and Arnc e Arr are the peak area of reagent in the was filtered by removing all MS/MS peaks within +/− 17 Da of the
negative control and in the reaction, respectively. precursor m/z. MS/MS spectra were window filtered by choosing only
the top 6 peaks in the +/− 50 Da window throughout the spectrum.
%C = 100 100x (Arr xAisnc )/(Arnc xAisr ) (1)
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
2.4. In vivo assays create consensus spectra. Further, consensus spectra that contained less
than 2 spectra were discarded. A network was then created where edges
For in vivo assays, mature oranges (Citrus sinensis) purchased from a were filtered to have a cosine score above 0.5 and more than 4 matched
local grocery store (Campinas, SP, Brazil) were surface sterilized and peaks. Further edges between two nodes were kept in the network only
wounded following the procedure described by Costa et al. (2019b) if each of the nodes appeared in each other's respective top 10 most
[14]. Oranges were infected with 15 µL of P. italicum or P. digitatum similar nodes. The spectra in the network were then searched against
spore solution (106 spores mL−1). Infected and control oranges were GNPS' spectral libraries. The library spectra were filtered in the same
2
93
J.H. Costa, et al. Bioorganic Chemistry 102 (2020) 104126
manner as the input data. All matches kept between network spectra After 6 h, only the aglycone flavonoid 3 was detected for both fungi
and library spectra were required to have a score above 0.5 and at least (Fig. S8 A and B). PE and PC also exhibited similar profile in the hy-
4 matched peaks [16]. 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
2.8. Imaging mass spectrometry (IMS) analysis assays. Lastly, GC did not show significantly enzymatic activity on 1,
with conversion rate close to zero during the experiment (Fig. S8 E).
For IMS, analyses were performed directly on orange peels infected PI exhibited the fastest and highest conversion rate of 4, reaching
by P. digitatum at 4 and 6 days post inoculation (dpi), following pre- 100% of the aglycone product after 48 h (Fig. S9 A), however, unlike
viously protocol [14]. A negative control was also performed. Analyses the hydrolysis of compound 1 to 3 that started at the initial time of the
were carried out using a desorption electrospray ionization (DESI) biocatalysis, the hydrolysis of flavanone 4 leading to compound 6
source Prosolia Model Omni Spray 2D®-3201) coupled to Thermo Sci- started at 3 h with 39% of conversion. PD showed another profile for
entific QExactive® Hybrid Quadrupole-Orbitrap Mass Spectrometer. the hydrolysis of 4, with a conversion rate close to 0% at the first
Methanol flow rate was 10.0 mL min−1. Mass resolving power of monitored time which was gradually increasing to 91% at 168 h (Fig.
70.000 at m/z 200 was used to acquire IMS data. Bin width of Δm/ S9 B), unlike what was observed for 1, where PD reached 100% of
z ± 0.03 was used to generated images at Firefly data conversion hydrolysis. Furthermore, the hydrolysis of 4 to 5 was only detected for
software (version 2.1.05). Images were visualized using Biomap soft- this fungus. PC reached 49% of hydrolysis of 4 at 168 h (Fig. S9 C) and
ware (version 3.8.0.4) developed by Novartis Institutes for BioMedical similar profile was observed for the hydrolysis of 1. GC and PE showed
Research, and color scaling was fixed during the processing of each a conversion rate for 4 close to 5% (Fig. S9 D and E), therefore, GC was
image. MS spectra were processed using Xcalibur (version 3.0.63) de- not able to effectively hydrolyze any of the flavanones, while PE only
veloped by Thermo Fisher Scientific. exhibited high enzymatic activity on compound 1.
The enzymes responsible for the hydrolyzing action observed in vitro
3. Results and discussion 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-
3.1. In vitro biocatalysis: Hydrolysis of citrus flavonoids rhamnosidase catalyzes the hydrolysis of hesperidin to hesperitin-glu-
coside or of naringin to prunin, releasing rhamnose. Subsequently, β-
Citrus flavonoids have been shown an important role during the glucosidase hydrolyzes hesperitin-glucoside into hesperidin or prunin
host-pathogen interaction [8,9,10,11,12]. As mentioned before, the into naringenin, releasing glucose [18,19,20,21,22,23]. In the litera-
flavonoids act as a resistance barrier produced by the host against the ture, some plants, bacteria, yeast and filamentous fungi are reported as
phytopathogens [8,9]; nevertheless, this defense mechanism is over- naringinase and hesperidinases producers, mainly fungi from the genus
come by the phytopathogens and the infections occur successfully in the Aspergillus and Penicillium. Fungal naringinases and hesperidinases are
plants. Therefore, studies that evaluate the interaction between the biotechnologically important and are applied in large amounts for in-
flavonoids and the phytopathogens are helpful to unravel the fungal dustrial uses; these enzymes are employed, by example, to reduce the
infection strategies and to understand the infection processes, leading bitterness of citrus fruit juice and to enhance wine aroma
to the development of methods for controlling postharvest diseases. [18,19,20,21,22,23]. Our data showed that P. italicum is a promising
In order to investigate the interaction of phytopathogens with the new source of naringinase and hesperidinase, since it exhibited faster
citrus flavonoids during infection process, biocatalysis assays were and higher hydrolytic activity when compared with other fungi.
employed using hesperidin and naringin, as substrates, and whole cells The hydrolyzing action of P. digitatum on citrus flavonoids were
of the main citrus pathogens and one non-citrus pathogen as catalysts. already known in the literature [10,12]. Furthermore, López-Peréz
The use of whole cells in biocatalysis has advantages compared to et al. (2015) [24] investigated P. digitatum genes expressed during in-
traditional chemical approaches as high selectivity, multi-step reactions fection in citrus by RNAseq. The authors reported that genes of hy-
in a single strain, cofactor regeneration, high catalytic efficiency and it drolase activities, such as hydrolysis of O-glycosyl compounds, were up-
is an environmentally friendly approach [17]. regulated during the infection process and the functions encoded by
Reactions were monitored by HPLC-MS analyses (Table S1) (Figs. these up-regulated genes are related to a pathogenic life style [24].
S1–S6) and the main biotransformations detected were the hydrolysis of However, although the hydrolyzing activity has an important role for
hesperidin (1) in hesperitin-7-O-glucoside (2) followed by formation of virulence, it is the first time that this enzymatic activity is measured for
hesperitin (3), and the hydrolysis of naringin (4) in pruning (5) and in P. digitatum and compared with other main citrus pathogens. The in
the respective aglycone naringenin (6) (Fig. 1). For PD and PI, causal vitro assays showed different profiles for the hydrolysis of naringin and
agents of the most incident diseases in citrus, the reactions were hesperidin by P. digitatum, which deserves future research to reveal how
monitored from 0 to 168 h (7 d), since we expected faster and higher this affect the infection process and host-pathogen interaction.
hydrolysis rate for these fungi. PC, GC and PD were monitored after 6 h. The hydrolysis activity detected for the most of the phytopathogens
The enzymatic conversions (%) of 1 and 4 by fungi cells were cal- contradicts data reported by previous studies, which observed that the
culated using Eq. (1) and are represented in Figs. 2 and 3, respectively. aglycons compounds have more antifungal action than the glycosylated
According to the graphs, obtained after monitoring the conversion rates flavanones [12]. Ortuño and Del Río (2009) [12] reported that 3 and 6
during 7 d, the phytopathogens have diverse biocatalytic pattern on are more active fungistatic agents against P. digitatum than 1 and 4. The
flavonoids, with some fungi reaching 100% of hydrolysis after 7 d concentration (mM) of 1, 3, 4 and 6, in PDA medium, to inhibit 50% of
compared to 0% of conversion by others. Furthermore, comparing the P. digitatum radial growth (IC50) was 7.86 ± 0.3, 0.12 ± 0.01,
hydrolysis activity for the same strain on the different substrates 1 and 17.9 ± 2.3 and 0.32 ± 0.02, respectively. To confirm this data, we
4, Figs. 2 and 3 also show different conversion profiles. performed antifungal assays using PDA medium supplemented with 1
PD and PI showed similar profiles in the enzymatic hydrolysis of 1, or 3 however, no significant difference was observed in fungal radial
with high conversion at the beginning of the biocatalysis assays and growth between 1 and 3 at the concentrations tested (50, 100 or
reaching 100% at 6 h. Apparently, PI demonstrated a higher conversion 400 mg L−1). (Fig. S10) which may indicate similar growth pattern in
rate of 1 when compared to PD at the first monitored time (77 versus the presence of both 1 and 3. Therefore, the difference in the antifungal
46%); however, until 6 h, most of the enzymatic product after hydro- potential between the flavonoids need to be revised and deserves fur-
lysis of 1 by PD was compound 3 (Fig. S7 A) while for PI the major ther investigation.
compound was the product 2 (Fig. S7 B), therefore, most of the enzy- In contrast to P. digitatum, G. citri-aurantii did not hydrolyze any of
matic activity by PI was the hydrolysis of the rhamnose portion of 1. the flavonoids, nevertheless the sour rot has been reported as a serious
3
94
J.H. Costa, et al. Bioorganic Chemistry 102 (2020) 104126
4
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, molecules of plant cell wall to facilitate their penetration [34]. β-glu-
although it hydrolyzed hesperidin. Furthermore, the authors only ob- cosidades are part of cell wall degrading enzymes (CWDEs) and during
served the hydrolysis activity and no transformation of aglycone fla- P. digitatum infection in citrus, CWDEs genes were up-regulated and
vonoids were detected [30]. These results are in agreement with our lacking mutants in genes enconding CDWEs were less virulent, sup-
data, since we did not detect any other enzymatic activity in vitro. porting the role in pathogenicity of specific CWDEs [24]. Nevertheless,
Ciegler et al. (1971) [30] attributes this inability to the fact that these the action of these enzymes on citrus defense metabolites, regarding to
flavonoids are relatively insoluble and impermeable into cells. virulence is not well established.
Naringinases e hesperidinases have not been related to the patho-
genicity in the literature. Penicillium species have been reported as
3.2. In vivo biotransformation assays
naringinase producers [31], however, a number of publications focus
on the immobilization of naringinases and application in industries
3.2.1. Hydrolysis of citrus flavonoids
[32]. Yet, Penicillium species also have been reported to have a high
For in vivo assays, PD and PI were selected since they presented
capacity to produce extracellular β-glucosidades [32,33]. Some β-glu-
higher conversion rate in vitro and are the main citrus pathogens re-
cosidades have been related as part of fungal virulence strategy, since
ported in the literature [2,4]. Similarly as the in vitro experiments, the
fungal pathogens secrete β-glucosidades to break the cellulose
main biotransformations observed were the hydrolysis of 1 into 2 and
5
96
J.H. Costa, et al. Bioorganic Chemistry 102 (2020) 104126
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- hydrolysis rate of flavonoids in vivo. Thus, while the fungus decreases
itored by HPLC-MS analysis at 3 and 5 days post inoculation (dpi). The the levels of 1 and 4, converting into the aglycones species 3 and 6
enzymatic conversation (%) of 1 and 4 was calculated using Equation respectively, the citrus fruit produce more 1 and 4, decreasing the en-
(1). zymatic conversion calculated.
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 3.2.2. Biotransformation of flavonoids in vivo by P. digitatum: A molecular
0.97% by PD and 0,90% by PI, while for 4, PD and PI hydrolyzed 2.08 network approach
and 3.82%, respectively. However, at the later infection, PI showed the To investigate in vivo hydrolysis of 1 and 4 and other bio-
highest conversion rates for the flavanones in vivo, when compared with transformation products present in this interaction, molecular net-
PD. At 5 dpi, PI hydrolyzed 6.43 and 17.08% of compounds 1 and 4, working was applied as a dereplication tool for fruit infected by P. di-
respectively, while PD exhibited conversion rate of 1.56% for com- gitatum to further characterize infection’s metabolic profile.
pound 1 and 2.56% for compound 4. Analyses of metabolite’s fragmentation patterns, acquired through
As mentioned before, flavonoids in citrus species act in the defense tandem MS techniques, and comparison with databases have become a
mechanism against fungi cells, and their composition can change standard procedure in natural product discovery and other metabolic
during pathogen attack [8,9]. In the in vitro study (section 3.1), it was studies [35]. In this study, we utilized GNPS platform (Global Natural
observed that PD and PI totally hydrolyzed the glycoside flavonoids in Products Social Molecular Networking), one of the largest natural
its aglycone forms (Figs. 2 and 3). Therefore, it was expected that in the products public database with regards to MS/MS spectra [16].
in vivo studies, this result would remain. Nevertheless, in vivo assays In molecular networking analyses, each node represents one meta-
showed that, although the enzymatic conversion had increase over bolite, labeled according to its m/z ratio. Based on similar fragmenta-
time, the percentage of this conversion was below expectation for both tion patterns, GNPS platform is able to group nodes that represent
PD and PI. This can be related to the fact that, while the fungus hy- molecules with similar structures in clusters [16]. GNPS dereplication
drolyzes the flavonoids, the fruit produces more compounds to defend with extracts from both PD infected and control orange fruit resulted in
itself from attack [8,9]. Kim et al. (2011) [9] measured the level of 1 a network containing 149 clusters. For the purpose of this study, clus-
and 4 in oranges infected by P. digitatum and reported that the levels of ters A, B and C (Fig. 4) were chosen for deeper analysis since they al-
these flavonoids decrease at the early stage of infection, gradually in- lowed visualization of the main citrus flavonoids. HRESI-MS data of all
crease in the middle stage and subsequent decrease at later infection. compounds reported in this study are compiled in Table S2.
The biosynthesis of the defense mechanisms by the fruit competes with Based on structure similarity, molecular network grouped in cluster
de fungus metabolism to consume them [9], which can explain the data A flavonoids bonded to disaccharides: hesperidin (1), naringin (4) and
observed here for the in vivo assays and that it is hard to measure the didymin (7), were identified through GNPS MS/MS library (Figs.
6
97
J.H. Costa, et al. Bioorganic Chemistry 102 (2020) 104126
S11–S13) and are flavonoids normally found in citrus fruit [36]. Cluster 4. Conclusions
B identified (Figs. S14–S16) and grouped products with a single sac-
charide: hesperitin-7-O-glucoside (2), pruning (5) and luteolin-7-O- Among the molecular factors that occur during the infection of ci-
glucoside (8), also compounds that can be commonly found in citrus trus fruit by phytopathogens, the involvement of the flavonoids is one
[36]. of the most known and yet not fully explored. As far as we know, this is
Interestingly, cluster C presented and identified the products of total the first report to study the enzymatic activity of phytopathogenic fungi
hydrolysis of 1 and 4: hesperitin (3) and naringenin (6) (Figs. S17 and on citrus flavanones. The in vitro biocatalysis assays employing nar-
S18). Also, other aglycone flavonoids such as isosakuranetin (9) (Fig. ingin, hesperidin and whole fungi cells demonstrated to be an appro-
S19), eriodictyol (10) (Fig. S20) and diosmetin (11) were grouped in priate technique to measure and detect enzymatic activities. According
this cluster. The presence of this compounds in the molecular network to our data, the hydrolysis activity is different among fungi, with P.
was expected, since they are present in citrus fruit [36]. However, digitatum and P. italicum showing high conversion rates and G. citri-
compounds 12 and 13 were only found in the infected fruit and were aurantii exhibiting no enzymatic activity. Also, through in vivo assays,
not identified as citrus flavonoids. Both possible isomers of 12 have molecular networking and IMS, our experiments demonstrated, for the
already been described in the literature, but never in citrus [37,38]. first time, novel compounds involved in defense of citrus against phy-
Based on fragmentation profile of 12 and 13 and their comparison topathogens. Lastly, this study opens new opportunities of research
to 3 and 11, it was possible to confirm the presence of a hydroxyl group since our data add more information about the fungal response against
in 12 and 13 in the A ring of the flavanone structure (Fig. 5). Flava- the plant defense molecules, and the enzymes described here are
nones present a common fragmentation profile containing retro- commercial and biotechnologically important.
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 Funding
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 This work was supported by the Coordenação de Aperfeiçoamento
difference of these fragments suggests the presence of a hydroxyl group de Pessoal de Nível Superior - Brasil (CAPES) [Finance Code 001],
in ring A of 12 and 13, however the correct position of the hydroxyl Fundação de Amparo a Pesquisa no Estado de São Paulo [grant numbers
group cannot be determined by mass spectrometry. 2018/13027–8, 2019/06359–7, 17/24462–4] and L’Oréal Brazil, to-
As mentioned before, flavonoids 12 and 13 were only present in PD gether with ABC and UNESCO in Brazil.
infected oranges, suggesting that these molecules are biotransformation
products of other citrus flavonoids. Based on this data, two hypotheses
CRediT authorship contribution statement
are possible: biotransformation is carried out by either (i) PD as an
infection mechanism or (ii) citrus as a defense mechanism.
Jonas Henrique Costa: Conceptualization, Investigation, Writing -
Flavonoids 12 and 13 were not found in the in vitro assays, in-
original draft. Laura Soler Fernandes: Investigation, Writing - original
dicating that their production may be part of citrus’ defense mechan-
draft. Daniel Yuri Akiyama: Investigation, Writing - original draft.
isms. Furthermore, based on structure similarity, 12 and 13 are the
Taícia Pacheco Fill: Conceptualization, Methodology, Resources,
hydroxylation products of 11 and 3, respectively. Studies analyzing P.
Supervision, Writing - review & editing.
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 Declaration of Competing Interest
genes that participate in the flavonoid biosynthesis pathway [40]. Thus,
we hypothesize that 12 and 13 are citrus biotransformation products as The authors declare that they have no known competing financial
a defensive mechanism to P. digitatum’s infection. interests or personal relationships that could have appeared to influ-
To confirm our hypothesis, IMS analyses were performed in oranges ence the work reported in this paper.
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
Acknowledgements
surface, correlating molecules and their spatial distribution [41]. In
previous studies, IMS technique was successfully applied in oranges
We would like to thank Dr. Katia Kupper for P. citrinum and
with green mold disease to monitor the infection process and the pro-
Geotrichum citri-aurantii strains, and Dr. Ljubica Tasic for the commer-
duction of secondary metabolites by P. digitatum [14]. Fig. 6 shows IMS
cially flavonoids. The authors thank Coordenação de Aperfeiçoamento
signals obtained for ion [M + H]+ 319.0813 (Fig. S26), which corre-
de Pessoal de Nível Superior - Brasil (CAPES) [Finance Code 001],
spond to compound 13 (error = 0.2 ppm). IMS images revealed that
Fundação de Amparo a Pesquisa no Estado de São Paulo [grant numbers
compound 13 is accumulated in the infected oranges when compared to
2018/13027–8, 2019/06359–7, 17/24462–4] and L’Oréal Brazil, to-
the control, yet, 13 is intensively concentrated on the peel about to be
gether with ABC and UNESCO in Brazil for the funding.
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- Appendix A. Supplementary material
pathogens and further investigation is needed to describe the role of
these compounds in citrus defense. Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bioorg.2020.104126.
7
98
J.H. Costa, et al. Bioorganic Chemistry 102 (2020) 104126
References [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.
[1] Citrus: World Markets and Trade. United States Department of Agriculture. Foreign [22] J. Wang, Y. Ma, X. Wu, L. Yu, R. Xia, G. Sun, F. Wu, Selective hydrolysis by com-
Agricultural Service, January, p. 1–13, 2020. Available online: https://downloads. mercially available hesperidinase for isoquercitrin production, J. Mol. Catal. B
usda.library.cornell.edu/usda-esmis/files/w66343603/00000g55g/kp78h0193/ci- Enzym. 81 (2012) 37–42, https://doi.org/10.1016/j.molcatb.2012.05.005.
trus.pdf (accessed 14 May 2020). [23] A.F.M. Furtado, M.A.P. Nunes, M.H.L. Ribeiro, Hesperidinase encapsulation to-
[2] I. Talibi, H. Boubaker, E.H. Boudyach, A.B. Aoumar, Alternative methods for the wards hesperitin production targeting improved bioavailability, J. Mol. Recognit.
control of postharvest citrus diseases, J. Appl. Microbiol. 117 (1) (2014) 1–17, 25 (2012) 595–603, https://doi.org/10.1002/jmr.2224.
https://doi.org/10.1111/jam.12495. [24] M. López-Pérez, A. Ballester, L. González-Candelas, Identification and functional
[3] M. El-Otmani, A. Ait-Oubahou, L. Zacarías, Citrus spp.: Orange, mandarin, tan- analysis of Penicillium digitatum genes putatively involved in virulence towards ci-
gerine, clementine, grapefruit, pomelo, lemon and lime, Postharvest Biol. Technol. trus fruit, Mol. Plant Pathol. 16 (3) (2015) 262–275, https://doi.org/10.1111/mpp.
Tropical Subtropical Fruits 437–514 (2011) 515e–516e, https://doi.org/10.1533/ 12179.
9780857092762.437. [25] Y. Wu, X. Duan, G. Jing, O.Q. Yang, N. Tao, Cinnamaldehyde inhibits the mycelial
[4] J.M. Bazioli, J.R. Belinato, J.H. Costa, D.Y. Akiyama, J.G.M. Pontes, K.C. Kupper, growth of Geotrichum citri-aurantii and induces defense responses against sour rot in
F. Augusto, J.E. Carvalho, T.P. Fill, Biological control of citrus phytopathogens, citrus fruit, Postharvest Biol. Technol. 129 (2017) 23–38, https://doi.org/10.1016/
Toxins 11 (2019) 460, https://doi.org/10.3390/toxins11080460. j.postharvbio.2017.03.004.
[5] M. Ramón-Carbonell, P. Sánchez-Torres, Significance of 195 bp-enhancer of [26] B. Deng, W. Wang, L. Deng, S. Yao, J. Ming, K. Zeng, Comparative RNA-seq analysis
PdCYP51B in the acquisition of Penicillium digitatum DMI resistance and increase of of citrus fruit in response to infection with three major postharvest fungi,
fungal virulence, Pestic. Biochem. Physiol. (2020), https://doi.org/10.1016/j. Postharvest Biol. Biotechnol. 146 (2018) 134–146, https://doi.org/10.1016/j.
pestbp.2020.01.003. postharvbio.2018.08.012.
[6] J.H. Costa, J.M. Bazioli, J.G. Ponte, T.P. Fill, Penicillium digitatum infection me- [27] J.A. Del Río, M.C. Arcas, O. Benavente-García, A. Ortuño, Citrus polymethoxylated
chanisms in citrus: What do we know so far? Fungal Biol. 123 (8) (2019) 584–593, flavones can confer resistance against Phytophthora citrophthora, Penicillium digi-
https://doi.org/10.1016/j.funbio.2019.05.004. tatum, and Geotrichum species, J. Agric. Food. Chem. 46 (10) (1998) 4423–4428,
[7] K. Shirasu, R. Maor, The arms race continues: battle strategies between plants and https://doi.org/10.1021/jf980229m.
fungal pathogens, Curr. Opin. Microbiol. 8 (4) (2005) 399–404, https://doi.org/10. [28] L. Díaz, J.A. Del Río, M. Pérez-Gilabert, A. Ortuño, Involvement of an extracellular
1016/j.mib.2005.06.008. fungus laccase in the flavonoid metabolism in Citrus fruits inoculated with
[8] A. Ortuño, L. Díaz, N. Alvarez, I. Porras, A. García-Lidón, J.A. Del Río, Comparative Alternaria alternata, Plant Physiol. Biochem. 89 (2015) 11–17, https://doi.org/10.
study of flavonoid and scoparone accumulation in different Citrus species and their 1016/j.plaphy.2015.02.006.
susceptibility to Penicillium digitatum, Food Chem. 125 (1) (2011) 232–239, https:// [29] M.T. Lafuente, A. Ballester, L. González-Candelas, Involvement of abscisic acid in
doi.org/10.1016/j.foodchem.2010.09.012. the resistance of citrus fruit to Penicillium digitatum infection, Postharvest Biol.
[9] H.G. Kim, G. Kim, J.H. Lee, S. Park, W.Y. Jeong, Y.H. Kim, S.T. Kim, Y.A. Cho, Technol. 154 (2019) 31–40, https://doi.org/10.1016/j.postharvbio.2019.04.004.
W.S. Lee, S.J. Lee, J.S. Jin, S.C. Shin, Determination of the change of flavonoid [30] A. Ciegler, L.A. Lindenfelser, G.E.N. Nelson, Microbial transformation of flavonoids,
components as the defence materials of Citrus unshiu Marc. fruit peel against Appl. Microbiol. 22 (1971) 974–979.
Penicillium digitatum by liquid chromatography coupled with tandem mass spec- [31] M.H. Ribeiro, Naringinases: occurrence, characteristics, and applications, Appl.
trometry, Food Chem. 128 (1) (2011) 49–54, https://doi.org/10.1016/j.foodchem. Microbiol. Biotechnol. 90 (6) (2011) 1883–1895, https://doi.org/10.1007/s00253-
2011.02.075. 011-3176-8.
[10] M.C. Arcas, J.M. Botía, A.M. Ortuño, J.A. Del Rio, UV irradiation alters the levels of [32] G. Molina, E.A. Lima, G.P. Borin, M.C.S. Barcelos, G.M. Pastore, Beta-glucosidase
flavonoids involved in the defence mechanism of Citrus aurantium fruits against from Penicillium, New and Future Developments in Microbial Biotechnology and
Penicillium digitatum, Eur. J. Plant Pathol. 106 (2000) 617–622. Bioengineering: Penicillium System Properties and Applications, Elsevier, 2018, pp.
[11] A. Ortuño, A. Baídez, P. Gómez, M.C. Arcas, I. Porras, A. García-Lídon, J.A. Del Río, 137–151, , https://doi.org/10.1016/B978-0-444-63501-3.00007-7.
Citrus paradisi and Citrus sinensis flavonoids: Their, influence in the defence me- [33] D. Mamma, E. Kalogeris, D.G. Hatzinikolaou, A. Lekanidou, D. Kekos, B.J. Macris,
chanism against Penicillium digitatum, Food Chem. 98 (2) (2006) 351–358, https:// P. Christakopoulos, Biochemical characterization of the multi-enzyme system pro-
doi.org/10.1016/j.foodchem.2005.06.017. duced by Penicillium decumbers grown on rutin, Food Biotechnol. 18 (1) (2004)
[12] A. Ortuño, J.A. Del Río, Role of Citrus phenolic compounds in the resistance me- 1–18, https://doi.org/10.1081/FBT-120030382.
chanism against pathogenic fungi, Tree Forestry Sci. Biotechnol. 3 (Special Issue 2) [34] P.K. Sudheeran, R. Ovadia, O. Galsarker, I. Maoz, N. Sela, D. Maurer,
(2009) 49–53. O. Feygenberg, M.O. Shamir, N. Alkan, Glycosylated flavonoids: fruit’s concealed
[13] J.H. Costa, B.Z. Costa, D.A. Angelis, A.J. Marsaioli, Monoamine oxidase and antifungal agents, New Phytol. 225 (2019) 1788–1798, https://doi.org/10.1111/
transaminase screening: biotransformation of 2-methyl-6-alkylpiperidines by nph.16251.
Neopestalotiopsis sp. CBMAI 2030, Appl. Microbiol. Biotechnol. 101 (2017) [35] B.C. Covington, J.A. McLean, B.O. Bachmann, Comparative mass spectrometry-
1061–1070, https://doi.org/10.1007/s00253-017-8389-z. based metabolomics strategies for the investigation of microbial secondary meta-
[14] J.H. Costa, J.M. Bazioli, E.V. Araújo, P.H. Vendramini, M.C.F. Porto, M.N. Eberlin, bolites, Nat. Prod. Rep. 34 (1) (2017) 6–24, https://doi.org/10.1039/c6np00048g.
J.A. Souza-Neto, T.P. Fill, Monitoring indole alkaloid production by Penicillium di- [36] G. Gattuso, D. Barreca, C. Gargiulli, U. Leuzzi, C. Caristi, Flavonoid composition of
gitatum during infection process in citrus by Mass Spectrometry Imaging and mo- citrus juices, Molecules 12 (8) (2007) 1641–1673, https://doi.org/10.3390/
lecular networking, Fungal Biol. 123 (8) (2019) 594–600, https://doi.org/10.1016/ 12081641.
j.funbio.2019.03.002. [37] D.C. Albach, R.J. Grayer, S.R. Jensen, F. Özgökce, N.C. Veitch, Acylated flavone
[15] J. Smedsgaard, Micro-scale extraction procedure for standardized screening of glycosides from Veronica, Phytochemistry 64 (7) (2003) 1295–1301, https://doi.
fungal metabolite production in cultures, J. Chromatogr. 760 (1997) 264. org/10.1016/j.phytochem.2003.08.012.
[16] M. Wang, et al., Sharing and community curation of mass spectrometry data with [38] A. Lenherr, M.F. Lahloub, O. Sticher, Three flavonoid glycosides containing
global natural products social molecular networking, Nat. Biotechnol. 34 (2016) acetylated allose from Stachys Recta, Phytochemistry 23 (10) (1984) 2343–2345,
828–837 PMID: 27504778. https://doi.org/10.1016/S0031-9422(00)80548-8.
[17] Y. Tao, B. Lin, Whole-cell biocatalysts by design, Microb. Cell Fact. 16 (2017) 106, [39] D. Tsimogiannis, M. Samiotaki, G. Panayotou, V. Oreopoulou, Characterization of
https://doi.org/10.1186/s12934-017-0724-7. flavonoid subgroups and hydroxy substitution by HPLC-MS/MS, Molecules 12 (3)
[18] Y. Zhu, H. Jia, M. Xi, J. Li, L. Yang, X. Li, Characterization of naringinase from (2007) 593–606, https://doi.org/10.3390/12030593.
Aspergillus oryzae 11250 and its application in the debitterization of orange juice, [40] L. González-Candelas, S. Alamar, P. Sanchez-Torres, L. Zacarias, J.F. Marcos, A
Process Biochem. 62 (2017) 114–121, https://doi.org/10.1016/j.procbio.2017.07. transcriptomic approach highlights induction of secondary metabolism in citrus
012. fruit in response to Penicillium digitatum infection, BMC Plant Biol. 10 (1) (2010)
[19] S. Yadav, R.S.S. Yadav, K.D.S. Yadav, An α-L-rhamnosidase from Aspergillus awamori 194, https://doi.org/10.1186/1471-2229-10-194.
MTCC-2879 and its role in debittering of orange juice, Int. J. Food Sci. Technol. 48 [41] S. Bertrand, N. Bohni, S. Schnee, O. Schumpp, K. Gindro, J. Wolfender, Metabolite
(2013) 927–933, https://doi.org/10.1111/ijfs.12043. induction via microorganism co-culture: A potential way to enhance chemical di-
[20] J. Bodakowska-Boczniewicz, Z. Garncarek, Immobilization of naringinase from versity for drug discovery, Biotechnol. Adv. 32 (6) (2014) 1180–1204, https://doi.
Penicillium decumbens on chitosan microspheres for debittering grapefruit juice, org/10.1016/j.biotechadv.2014.03.001.
Molecules 24 (23) (2019) 4234, https://doi.org/10.3390/molecules24234234.
8
99
Jonas Henrique Costaa, Laura Soler Fernandesa, Daniel Yuri Akiyamaa, Taícia
Pacheco Filla*
Supplementary data
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.
Figure S25. Oranges fruits (Citrus sinensis) analyzed by Imaging Mass Spectrometry (IMS).
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
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
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”
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
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.
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
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.
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.
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.
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
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.
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
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
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.
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).
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.
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/.
SCIENTIFIC REPORTS
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
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 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
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. 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 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).
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 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
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 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
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
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
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
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
10 ANEXOS
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
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 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
DECLARAÇÃO