Colletotrichum gloeosporioides isolated from tropical fruits.
Resumo
Um total de 580 cepas de leveduras, isoladas a partir de frutos tropicais, foram avaliados quanto à sua capacidade para produzir a toxina killer e atuar como agentes de biocontrole in vitro contra o fitopatogeno Colletotrichum gloeosporioides. Destas, 29 foram capazes de expressar o fenótipo killer, sendo em seguida avaliadas quanto à sua capacidade de controlar C. gloeosporioides in vitro. Todas as cepas killer foram identificadas pelo sequenciamento da região D1/D2 do 28S rRNA, em que ficou demonstrada a presença de Candida aaseri, Wickerhamomyces anomalus, Pichia kluyveri, Meyerozyma guilliermondii, Kodamaea ohmeri,. Cinco leveduras foram capazes de inibir em 100% a germinação de conídios em meio líquido e reduzir o crescimento micelial, em meio sólido de Colletotrichum gloeosporioides in vitro, com destaque para M. guilliermondii (cepa 443) que foi capaz de reduzir o crescimento micelial do fitopatogeno em 60% em meio sólido. Culturas filtradas e autoclavadas não apresentaram efeitos sobre o crescimento do patógeno. Estes resultados indicam o uso potencial de leveduras killer isoladas a partir de frutas tropicais como agentes de biocontrole de C. gloeosporioides em mamoeiro.
Isolation, identification and activity in vitro of killer yeasts against Colletotrichum
gloeosporioides isolated from tropical fruits.
Jaqueline Rabelo de Limaa, Luciana Rocha Barros Gonçalves c, Luciana Rocha Brandãod, Carlos
Augusto Rosad, Francisco Marto Pinto Vianab*
aPrograma de pós-graduação em Biotecnologia -RENORBIO – Universidade Estadual do Ceará. Empresa Brasileira de Pesquisa Agropecuaria - EMBRAPA- Laboratório de Patologia Pós-colheita, Rua Dra. Sara Mesquita, 2270, Planalto do Pici, CEP 60511-110. Fortaleza-CE, Brasil.
b Empresa Brasileira de Pesquisa Agropecuaria - EMBRAPA Agroindústria Tropical- Laboratório de Patologia Pós-colheita, Rua Dra. Sara Mesquita, 2270, Planalto do Pici, CEP 60511-110. Fortaleza-CE, Brasil.
c Departamento Engenharia Química- Universidade Federal do Ceará - UFC- Campus do PICI, CEP: 60455 – 760, Fortaleza-CE, Brasil.
dDepartamento de Microbiologia, ICB, CP 486, Universidade Federal de Minas Gerais, Belo Horizonte, MG 31270-901, Brazil.
* Corresponding author at: EMBRAPA Agroindústria Tropical- Laboratório de patologia pós-colheita Rua Dra. Sara Mesquita, 2270, Planalto do Pici, CEP 60511-110. Fortaleza-CE, Brasil. Tel.: +55 085 3391 7264 - Fax: +55 085 33917222 - E-mail address: jaquerabelo@hotmail.com (Jaqueline Rabelo de Lima a).
Abstract
A total of 580 yeasts strains, isolated from Ceara State of Brasil, were evaluated for their ability to produce killer toxin. Of these strains, 29 tested positive for the killer phenotype and were further evaluated for their ability to control Colletotrichum gloeosporioides germination in vitro. All yeast strains that expressed the killer phenotype were characterized by sequencing the D1/D2 regions of the large subunit of the rRNA gene. Five yeast strains provided a significant reduction in mycelial growth and conidial germination of C. gloeosporioides in vitro, especially Meyerozyma guilliermondii, which was able to reduce the fungal mycelial growth on solid medium (PDA) by 60% and block 100% of conidia germination in liquid media (PDB). Filtering and autoclaving the liquid cultures had no effect on the growth of the pathogen. These results indicate the potential use of antagonist yeasts isolated from tropical fruits in the control of anthracnose caused by C. gloeosporioides in papaya. Further elucidation of main mechanisms involved on anthracnose control by these yeasts could be helpful for the development of biocontrol techniques related to the management of this disease in tropical fruits.
5.1 Introduction
Anthracnose caused by Colletotrichum species is the main disease in post-harvest fruits and is considered to be a disease of high economic importance in Brazil's Northeast region [1]. Fungicides are widely used as pre- or postharvest treatments and are the main approach for reducing fruit losses from anthracnose. However, the use of fungicides has been increasingly restricted because of public concerns about toxic residues as well as consumer demands for less pesticide residues in foodstuffs [2, 3]. Moreover, the increasing emergence of pathogens that are resistant to fungicides reinforces the need to search for alternative methods for post-harvest disease control [4, 5, 6].
Biological control by antagonistic microorganisms, including yeasts, yeast-like fungi and bacteria, appears to be particularly promising as an approach to prevent fungal infections in different fruits and vegetables [7, 8, 9, 10]. The potential use of yeasts for the biocontrol of post-harvest diseases has been reported by several authors [3,11, 12, 13, 14,15, 16, 17, 18].
Yeasts have many features that make them effective as biocontrol agents in fruits and other foods. They can grow rapidly on inexpensive substrates in fermenters and are therefore easy to produce in large quantities [19]. In addition, they do not produce allergic spores or mycotoxins, in contrast to filamentous fungi [20], and they have simple nutritional requirements that enable them to colonize dry surfaces for long periods of time [21]. Buck [22] reported that saprophytic fungi are a common component of the aerial plant surface mycoflora and suggested that these microorganisms provide a natural antagonistic barrier against fungal plant diseases.
Special attention has been given to yeasts that exhibit the killer phenotype that was first described in 1963. This killer activity can be defined as the ability of some yeast species to secrete proteins or glycoprotein killer toxins that generally kill susceptible cells [23].
When the killer phenotype was first described, it was believed that these yeasts were only able to kill other yeasts; however, in recent years, several studies have reported that these microorganisms also have activity against filamentous fungi [16, 23, 24, 25].
The introduction of biocontrol agents for commercial use has encountered many obstacles, such as the consumer fear that biological agents or toxins will enter the diet, but most post- harvest biocontrol agents were originally isolated from fruit and vegetables and are indigenous to agricultural commodities [5]. Moreover, even when these biocontrol agents are introduced in large numbers to the surface of a commodity, they can only survive and grow in very restricted sites on the fruit surface (e.g., surface wounds). After the introduction of these agents to intact fruit surfaces, antagonist populations usually diminish to the level of natural epiphytic microbiota within a very short period of time. Moreover, yeasts have been used in fermentation processes for thousands of years [5, 13, 23], which reinforce their safety for human consumption purposes. The aim of this study was to isolate and identify yeasts expressing the killer phenotype from tropical fruits as potential biocontrol agents of Colletotrichum gloeosporioides.
5.2 Materials and Methods 5.2.1. Fruits harvest
From August 2009 to May 2010, 87 samples of tropical fruits (papaya, Carica papaya L.; cashew, Anacardium occidentale L.; sapoti, Manilkara zapota L.; mango, Mangifera indica; murici, Byrsonima crassifolia L. Rich and acerola, Malpighia glabra L.)
were collected from street markets and producer farms in the Ceará State, Brazil. The samples were packed in sterile plastic trays and immediately transported to the Plant Pathology Laboratory of Empresa Brasileira de Pesquisa Agropecuária- EMBRAPA, located in Fortaleza city, Ceará State, northeast Brazil.
5.2.2. Yeasts isolation
The isolation of yeasts was performed by weighing assepticaly aliquots of 25 g of peel and skins from each fruit and diluting them in 225 mL of sterile saline solution to generate a 10-1 dilution. From these samples, serial dilutions were obtained in sterile saline solution (10-2 and 10-3). An aliquot of 0.1 ml of each dilution was spread on a plate of potato dextrose agar (PDA) (Difco), pH 3.5, adjusted with sterile tartaric acid (10%) and incubated at 28 °C for 72 hours. This procedure was peformed in duplicate. After the incubation period, 5 colonies from each plate were selected based on different morphological characteristics. The yeast isolates were re-streaked on PDA to obtain pure cultures and stored at 4 °C for further studies.
5.2.3. Killer activity determination
Killer activity was tested in triplicate on YEPD (yeast extract 10 g L-1(Difco), peptone 20 g L-1(Difco), glucose 20 g L-1(Difco), agar 25 g L-1) with 0.003% of methylene blue and buffering with 0.01 M citrate buffer to pH 4.2. The strains Candida glabrata Y-55 (NCYC 388) and Saccharomyces cerevisiae (NCYC 1006), which are sensitive to most known mycocins, were grown for 24 h at 28 ºC on YM agar and then suspended in sterile distilled water at 4 x 105 cfu/mL and spread on the medium surface with sterile swabs [26].
Yeast strains to be tested for mycocin production were grown as previously described and inoculated in streaks on medium that had been inoculated with the sensitive strain. The plates were incubated at 28 ºC and observed daily for 3 days. Isolates were considered mycocinogenic if they produced an inhibition zone with no growth and adjacent blue zones that indicated the cellular death of the sensitive strain. Strains of S. cerevisiae (NCYC 232) and S. cerevisiae (NCYC 732) were employed as positive controls known to have mycocinogenic activity [26].
5.2.4. Determination of the optimal pH and temperature conditions for killer phenotype expression
Each killer yeast strain was evaluated using YEPD agar (yeast extract 10 g L-1
(Difco), peptone 20 g L-1(Difco), glucose 20 g L-1(Difco), agar 25 g L-1) with 0.003%
methylene blue.To adjust the different pH values (3.5, 4.0, 4.5, 5.0 and 5.5), a 0.01M citrate- phosphate buffer was used. For this test, susceptible S. cerevisiae (NCYC 1006) and C. glabrata (NCYC 388) strains were grown on PDA at 25° C for 24 hours and suspended in a sterile saline solution at a concentration of approximately 4.0 x 105 cells mL-1, spread with sterile swabs onto the surface of media and incubated at 25 °C for 30 minutes. The killer yeast strains to be evaluated have been inoculated onto agar surface, as a circle point, with, approximately, 1 cm of diameteron the surface of plates with the sensitive yeast strains. The plates were incubated at 20 °C, 25 °C, 30 °C and 35 °C for 72 hours. The presence of an inhibition zone with precipitation around the inoculation site was considered a positive test. Media conditions under pH 3.5 did not demonstrate adequate growth of the susceptible strains; thus, the test was discarded.
5.2.5. Morphological and physiological characteristics
The morphological characteristics of the yeast antagonists were examined by cell and colony patterns observation using the method described by Kurtzman [27]. Colony morphology of each yeast isolate was examined in cultures grown in solid medium (5% malt extract agar) at 25 °C. The features of the yeast cultures on the plates, i.e., color and texture, were recorded after 3 to 7 days of incubation. Yeasts physiological characteristics were determined according to their ability to ferment certain sugars semi-anerobically and to assimilate various carbon compounds as the major source of carbon under aerobic condition. Morphological and physiological characteristics were used for the confirmation of the yeasts identification by the rDNA sequencing technique. Physiological characteristics are presented in Table 5.1.
5.2.6. Yeasts identification
All killer toxin- producing yeasts were preliminarily grouped based on cultural morphology, urease production and physiological characteristics, determined by assimilation tests of carbon and nitrogen sources and the production of amyloid compounds [28]. Isolates with identical morphological and physiological characteristics were grouped together and subjected to PCR fingerprinting using the microsatellite-primed PCR technique (MSP-PCR) employing a synthetic oligonucleotide (GTG)5.
5.2.6.1. DNA extraction
For DNA extraction, yeast colonies were grown on modified Sabouraud agar (glucose 20 g L1(Difco), peptone 10 g L-1(Difco), yeast extract 5 g L-1(Difco), and agar 20 g L-1(Difco)) at 15 °C overnight, transferred to 2 mL sterile tubes (Eppendorf) with 100 µl of extraction buffer solution (50 mmol Tris L-1, 250 mmol NaCl l-1, 50 mmol EDTA l-1, 0.3% w/v SDS, at pH 8), and incubated at 65 °C for 30 min. Afterwards, 100 µl of phenol/chloroform/isoamilic alcohol (25:24:1) was added. The mixtures were vigorously vortexed, incubated for 3 min and centrifuged for 3 min at 1,5115 g. The DNA was dried overnight at room temperature, suspended in 100 mL of TE buffer (10 mM Tris, 10 mM Na- EDTA, at pH 8.0) and stored in a refrigerator at 4 °C [29].
5.2.6.2. PCR fingerprinting
PCR fingerprinting was performed in a 25-μL reaction containing 2.5 μL of 10X Mg-free PCR buffer, 1.5 μL of 25 mM MgCl2, 1 μL of 2 mM deoxyribonucleotide triphosphates (dNTPs), 1 μL of each (GTG)5 primer at 10 pmol, 5 μL of DNA template and
0.2 μL of 1 U/μL Taq DNA polymerase (Fermentans, USA). The PCR reactions were performed according to Libkind et al. [30]. Yeast strains with identical DNA banding patterns were grouped and categorized as putatively belonging to the same species [31]. At least one representative strain of each MSP-PCR group was subjected to sequence analysis of the D1/D2 domains of the large subunit of the rRNA gene as described below. Physiologically distinct strains with unique MSP-PCR banding patterns were also selected for direct identification by sequencing of the D1/D2 region of the rRNA gene.
5.2.6.3. Sequencing analysis
Total DNA was extracted using the methods described above. Cycle sequencing of the 600–650 bp region at the 5‘-end of the 26S rRNA D1/D2 domain was performed using the forward primer, NL1 (5‘-GCA TAT CAA TAA GCG GAG GAA AAG), and reverse primer, NL4 (5‘-GGT CCG TGT TTC AAG ACG G), as previously described by Lachance et al. [32].
The amplified DNA was concentrated and purified using Wizard Plus SV columns (Promega, USA) and sequenced in a MegaBACETM 1000 automated sequencing system (GE Healthcare, USA). The sequences obtained were compared to those included in
the GenBank database using the Basic Local Alignment Search Tool (BLAST, http://www.ncbi.nlm.nih.gov) [33]. The GenBank accession numbers for the sequences derived in this study are shown in Table 5.1.
5.2.7. Evaluation of the antagonistic properties of the yeast isolates against C.
gloeosporioides in solid medium
The antagonism in vitro in solid medium was evaluated according to the methodology described by Rosa et al.[18], with minor modifications. Discs with active C. gloeosporioides mycelium (5 mm in diameter and 10 days old) were deposited onto the center of the PDA plates. The yeast samples to be tested were then streaked onto the plates at two sites adjacent to the C. gloeosporioides discs, 3.5 cm away from the centered disc. Negative control consisted of PDA plates solely with C. gloeosporioides discs inoculation. For each treatment (each yeast strain), five experimental units (plates) were used. The plates were incubated at 28 °C, alternating between 12 h light/12 h darkness. The growth of the phytopathogen was daily followed. The experiment was finished when C. gloeosporioides growth on the positive control units reached the plate borders. The average mycelium growth was subjected to analysis of variance (ANOVA) and, when an overall difference was observed, the mean differences were compared using the Tukey‘s test with a 5% level of probability. Each experiment was repeated twice.
5.2.8. Antagonist action of yeasts on the germination of C. gloeosporioides spores in solid medium
Yeast strains inoculum was prepared using the following procedure: An inoculum loop with yeast cells was transferred to four flasks containing 20 mL of potato dextrose broth (PDB, Difco) and incubated at 28 °C for 24 h. One flask of a 20-mL culture was then filtered through a 0.22-μm membrane (T1). Another 20 mL of the culture was autoclaved at 121°C for 30 min.(T2). A non-washed culture constituted T3 and yeasts suspensions of 106 (T4), 107 (T5) and 108 (T6) cfu/mL were prepared. Senerade® - Bacillus subtilis (8 ppm) and Recop® - copper oxychloride (200 ppm) were used as positive controls (T7 and T8, respectively), and sterile distilled water was used as a negative control (T9).
For each treatment, 100 µL was added into a well (6 mm in diameter) in the center of a plate containing PDA plus 1.0 x 105 fungi spores/ml. All plates were incubated at 28 °C for ten days. The fungal growth was observed, and its colony diameter (average length of the
diameter in the x-axis and y-axis) was measured after the incubation period. Each treatment (each isolate) was conducted in triplicate, and each experiment was repeated twice. The yeast isolates able to completely inhibit fungal growth were considered to be antagonists and were kept for further studies.
5.2.9. Antagonist action of yeasts on the germination of C. gloeosporioides spores in liquid medium
The effect of five antagonist strains (M. guilliermondii 443 and W. anomalus 419, 420, 422 and 440) on C. gloeosporioides spores germination was assessed in PDA, as reported by Zhang et al.[17], with minor modifications. Aliquots of 100 µL of each treatment (previously described) were added to tubes containing 480 µL of PDB. Aliquots (100 µL) of C. gloeosporioides spores suspended in saline solution (5×106 spores/mL) were transferred to each tube. After 20 h of incubation at 28 °C, 100 spores/replicate were microscopically observed to determine the germination rate. Three tubes were used for each treatment, and three samples from each tube were evaluated. All of these experiments were repeated twice. A spore was considered germinated when the length of the germ tube exceeded the diameter of the spore, and the conidial germination percentage was estimated from the observation of 100 conidia on each slide [12].
5.2.10. Effect of antagonist yeasts filtrates on different C. gloeosporioides spores concentration germination
Filtered antagonist yeasts cultures (M. guilliermondii 443 and W. anomalus 419, 420, 422 and 440) were prepared as detailed in 2.7 section. Same assay described in 2.8 section was carried out, varying, however, C. gloeosporioides spores concentration (suspended in saline solution). Three different spores concentrations were employed: 5×103 spores/mL, (Treatment 1, t1), 5×104 spores/mL (Treatment 2, t2) and 5×105 spores/mL(Treatment 3, t3).
5.2.11. Statistical analysis
The data were analyzed using the analysis of variance (ANOVA) in the SISVAR software package (version 5.3; Universidade Federal de Larvas) [34]. Statistical significance was determined at the level P≤0.05. Replication averages with similar results were analyzed together. Average and standard error were calculated and reported for each experiment. When
averages were significantly different, a comparison was performed using Tukey‘s test (P ≤ 0.05).
5.3.0. Results
5.3.1. Yeast isolation, identification and killer activity determination
From August 2009 to May 2010, 580 yeast isolates were obtained from tropical fruit surfaces. Of the 580 yeast strains investigated, 29 isolates (5%) belonging to five different species produced an inhibition zone (halo) without cellular growth and a blue zone of cellular death, and these strains were considered to be killer yeasts. When the halo was evident but the blue zone of cellular death was not, the yeasts were considered antagonists and were not classified as killer strains.
PCR group fingerprints of the initial 29 killer yeasts isolates were performed and DNA banding patterns evidenced 17 different groups, from which, one representative strain was chosen for sequence analysis of the D1/D2 domains of the 28S rDNA to obtain an identification down to the species level. D1\D2 molecular identification of the 29 isolates showed the presence of five different species, including 17 strains of Kodamaea ohmeri (57%), 5 strains of Wickerhamomyces anomalus (17%), 4 strains of Pichia kluyvery (13%), 2 strains of Candida aaseri (7%) and 1 strain of Meyerozyma guilliermondii (3%). At least one representative strain of each identified species and all strains with robust results in the in vitro biocontrol tests (M. guilliermondii 443 and W. anomalus 419, 420, 422 and 440) were subjected to sequencing of the D1/D2 domains of the large subunit of the rRNA gene, and these sequences were deposited in GenBank (accession numbers are indicated in Table 5.1).
Table 5.1. Identification and physiological characteristics of yeasts isolated from tropical fruits. Code GenBank accession number Closest related species/GenBank accession number Similarity (%)
Identification Assimilation of nitrogen compounds
Physiological characteristics 10% NaCl a Glucose
c
lysine nitrate nitrite 5% 50%
63, 82, JN676081 Candida aaseri /U45802.1 99 Candida aaseri + - - + + -
150, 164,165,166,170, 449, 515, 534, 535, 567, 413, 522, 523, 524, 533, 539, 554
JN676082 Kodamaea /U45702.1 ohmeri 99 Kodamaea ohmeri
+ - - + + -
559, 88, 92, 95 JN676084 Pichia kluyveri /U75727.1 99 Pichia kluyveri + - - - + - 417, 419* JN627211 Pichia anomala / U74592 99 Wickerhamomyces anomalus + + + + + - 420* 27212 JN6 Pichia anomala / U74592 99 Wickerhamomyces anomalus + + + + + - 422* 27213 JN6 Pichia anomala / U74592 99 Wickerhamomyces anomalus + + + + + - 440* 27209 JN6 Pichia anomala / U74592 99 Wickerhamomyces anomalus + + + + + - 443* 27210 JN6 Pichia guilliermondii / U45709 100 Meyerozyma guilliermondii + + + + + -
* Antagonistic isolates used as biocontrol agents on C. gloesosporioides in in vitro tests.
5.3.2. Optimal conditions for the expression of the killer phenotype
The effects of incubation temperature and media pH on killer phenotype expression are shown in Figure 5.1. All isolates inhibited the growth of the susceptible strains when incubated at 25 °C and pH 4.5, which is in accordance with most other studies [35].
Fig. 5.1. The effect of incubation temperature and pH on killer phenotype expression in
various yeast strains.
5.3.3. Antagonist action of yeasts on the C. gloeosporioides mycelial growth on solid medium
Five isolates were found to possess an intense antagonism against the pathogen with a significant reduction of C. gloeosporioides mycelial growth when compared to control treatment (Fig. 5.2). These isolates belong to the species M. guilliermondii (strain 443) and W. anomalus (strains 419, 420, 422 and 440).
Fig. 5.2. The biocontrol efficacy of Meyerozyma guilliermondii 443 and Wickerhamomyces
anomalus 419, 420, 422 and 440 (applied at 108 cells/mL) in reducing the mycelium growth
of Colletotrichum gloeosporioides in solid medium. The results are the mean of two independent experiments. Standard error bars of the means are included. Values followed by the same letter are not statistically different by Tukey‘s test (p < 0.05).
5.3.4. Antagonist action of yeasts on the germination of C. gloeosporioides spores on solid medium
The strains 443, 419, 420, 422 and 440 reduced conidia germination in solid medium, and the strain 443 (M. guilliermondii) reduced germination as efficiently as the commercial culture of Senerade® - Bacillus subtilis (8 ppm). Autoclaved yeast cultures had no effect on conidia germination. Filtered cultures were unable to inhibit spores germination, even when lower C. gloeosporioides spores concentration were employed (105, 104 and 103 spores/ml). The commercial fungicide, copper oxychloride, was used as positive control and also had no effect on spores germination. The best results were obtained when washed cultures at a cellular concentration of 1.0 x 108 cells∕mL were employed (Fig. 5.3).
Fig. 5.3. Biocontrol efficacy of the five selected antagonist isolates (M. guilliermondii 443
and W. anomalus 419, 420, 422 and 440) applied at 106 cells/mL, 107 cells/mL, and 108