Biological control in the germination of seeds from two species native of the Cerrado region

February 28, 2021 (With 2 figures) Abstract Microorganisms have been efficiently used for the biological control of phytopathogens through the production of antimicrobial substances. However, the objectives of this work were: to study the germination of Butia purpurascens Glassman and Butia archeri Glassman seeds in different substrates, to select and identify the endophytic and rhizospheric bacterial isolates of B. purpurascens and B. archeri , and to perform antibiosis tests based on the isolated microorganisms of these tree species. No difference was found between the cultivation substrates for the percentages of germination, hard seeds, and fungal contamination in the B. purpurascens seeds. The Bacillus subtilis isolated showed the best capacity for suppressing the growth of the two deteriorative fungi tested in B. purpurascens seeds. No difference was found for inhibition of the growth of Aspergillus niger fungus (deteriorative fungus of B. archeri seeds) between the microorganisms with Bacillus sp. and Brevibacillus brevis compared to the control. In the microbiolization of B. purpurascens and B. archeri seeds performed with microbiological solutions produced from the endophytic and rhizospheric strains of Bacillus sp., no differences were observed in the percentages of germination and contamination by fungi. For B. archeri seeds, there was contamination by fungi and bacteria after one day of cultivation, primarily in the regions with lesions caused


Introduction
It is estimated that the Brazilian Cerrado contributes 10,000 plant species of the 60,000 phanerogams distributed across the country, marking this region as the richest in terms of flora diversity among the savannas of the world. It includes countless medicinal, ornamental, and fruit-bearing species that are important to the local populations (Gusmão et al., 2006) and is also considered one of the "hotspots" for global biodiversity conversation (Myers et al., 2000).
In this study, we used two plant species of great importance from the Cerrado region, Butia purpurascens Glassman and Butia archeri Glassman, both belongings to the family Arecaceae, popularly known as purple yatay palm and dwarf jelly palm, respectively. These species have great potential in ornamental, culinary applications owing to the flavor and fragrance of their fruits (Lorenzi et al., 2010), and in the therapeutic treatment of skin diseases (Hoffmann et al., 2014).
Species from the genus Butia are classified as critically endangered, with an estimated reduction in the population of over 80% over the next 10 years. As it is native to the southern region of Goias and the northern region of the Minas Gerais Triangle, it competes for land with soy and sugarcane plantations, further exacerbating the extinction process. Seeds of these species have a high rate of contamination by microorganisms during the pre-and post-emergence processes, resulting in the formation of abnormal seedlings and root rot (Magalhães et al., 2008;Bozza, 2009;IUCN, 2011).
Members of the family Arecaceae reproduce largely through sexual reproduction. Owing to the lengthy duration and low germination efficiency, phytoregulators must be applied and/or procedures must be adopted that facilitate the absorption of water, such as mechanical scarification. Although scarification can accelerate and increase the percentage of germination, the resulting wounds become an entry point for harmful microorganisms, which hinder or impede germination, resulting in a significant reduction in the number of viable plants (Rubio Neto et al., 2012. The use of specific microorganisms for the biological control of phytopathogens has proven to be efficient against certain fungi and bacteria owing to the production of antimicrobial substances (Rocha et al., 2009;Yuan et al., 2017;Saito et al., 2018;Liu et al., 2018), which leads to an alternative to the chemical control of diseases. These biological control agents can act directly and indirectly by inducing the plant's defense mechanisms in the control of fungal diseases (Benítez et al., 2004).
The interactions between pathogenic microorganisms and their antagonists include antibiosis, competition, parasitism, predation, and induction of the host's defenses (Melo and Azevedo, 1998). Most of the microorganisms involved in biological control operate by means of antibiosis, in which a metabolite produced by one microorganism has a harmful effect on the other, thereby inhibiting its growth. The production of metabolites may result in complete lysis and dissolution of the cellular structure, regardless of physical contact between the microorganisms (Remuska and Pria, 2007).
The objective of this study was to determine the best germination environment for B. purpurascens and B. archeri, identifying the deteriorative microorganisms infecting their seeds, and testing the ability of bacteria isolated from these tree species to control these microorganisms.

Obtaining the seeds
The study was carried out at the Federal Institute of Education, Science, and Technology of Goias -Rio Verde Campus, GO, in partnership with the Seeds, Plant Tissue Culture, and Agricultural Microbiology laboratories. The fruits of Butia purpurascens and Butia archeri were collected, respectively, 17°35'10.51"S, 50°59'13.06"W, 822m and 17°35'10.51"S, 50°59'13.06"W, 822m, in the city of Rio Verde, GO. The two species were intercropped with pasture and their fruits were randomly collected.
The fruits were pulped to obtain the diaspores and, subsequently, the seeds were extracted using a bench vise. The extracted seeds were disinfected in 70% alcohol for 1 minute and 40% sodium hypochlorite for 4 minutes. Next, they were washed in distilled water and autoclaved three times. Mechanical scarification was carried out in the laminar flow, removing the tegument in the hilar region (Rubio Neto et al., 2014). The phytosanitary and germination evaluations of the B. purpurascens and B. archeri seeds were performed according to the experimental tests described below.

Germination of B. purpurascens seeds in different substrates
Three germination substrates were evaluated: rolls of Germitest  paper, glass plates (120 × 150 mm) covered with washed and autoclaved sand, and Germitest  paper. On the plates containing sand, a field capacity of 60% was maintained weekly, while the paper was moistened with distilled autoclaved water in the amount of 2.5 times the weight of this dry substrate.
The seeds were kept in a Mangelsdorf germinator set at 30 °C for 30 days. The experiment used a completely randomized design, with seven repetitions of three germination substrates. The test was evaluated on a daily basis. When contamination by fungi or bacteria was detected, the contaminated seeds were removed from the germination environment and taken to the Laboratory of Agricultural Microbiology for identification and antibiosis tests. After 30 days, we evaluated the percentages of germination, contamination by fungi and bacteria, and hard seeds, the latter of which is characterized by the absence of contamination and germination. Only seeds with a cotyledon petiole of at least 1 cm in length were considered germinated.
Next, the data normality and the homogeneity of the variances were evaluated through the Shapiro-Wilk and Bartlett tests (5%), respectively, and the averages were compared by the Tukey's test (5%).

Selection of the endophytic and rhizospheric bacterial isolates of Butia purpurascens and Butia archeri
One hundred and fifty-three bacteria (67 endophytics from roots and 86 rhizospherics) were isolated from B. purpurascens and 110 bacteria were isolated from B. archeri (22 endophytics and 88 rhizospherics). Next, a qualitative test was performed on a Petri dish, containing PDA culture medium (infusion of 200 g of potatoes, 200 g of dextrose and 15 g of agar), in which four bacterial isolates were inoculated in the center of the plate opposite the fungal isolate. Bacteria that impeded the normal growth of the fungal mycelium were considered antagonistic when compared to the control plate, which was inoculated only with the fungal isolate.

Antibiosis test
To quantify the degree of antagonism of the seed-deteriorating microorganisms, we selected eight bacteria (seven rhizospherics and one endophytic) isolated from the species B. purpurascens, and two bacteria (one rhizospheric and one endophytic) from B. archeri.
The tests were performed according to the dual culture method (Mew and Rosales, 1986).
After the seed-deteriorating fungi were isolated, those that appeared with the greatest frequency were selected. For B. purpurascens, the fungal species Neodeightonia phoenicum BP91DF and Penicillium purpurogenum BP110DF were selected, while Aspergillus niger BA163D was isolated from B. archeri seeds. The fungi were multiplied in PDA culture medium for a period of seven days. Next, the fungal mycelium was inoculated in the center of the Petri dish. Neodeightonia phoenicum BP91DF and A. niger BA163DF were distanced 3 cm apart with immediate inoculation of the potential antagonists on opposite sides. The fungus P. purpurogenum BP110DF, owing to its slow growth, was inoculated 48 hours in advance, distanced 2 cm apart.
The experiment used a completely randomized design, with three repetitions for each fungus. The plates were incubated at room temperature for 4 to 8 days, depending on the speed of fungal growth. The diameter of the fungus was measured with a caliper, and the zone of fungal growth inhibition owing to the production of suppressive compounds by the bacteria was noted.

Identification of the microorganisms
The bacteria that expressed some capacity for antagonism and the seed-deteriorating fungi that were isolated were submitted for molecular identification. The bacteria were identified based on 16S rDNA, and the fungi were identified through partial sequencing of the internal transcribed spacer (ITS) from the rDNA region taken from one representative of each morphotype group.
The fungal isolates were grown in a liquid PD medium (400 mL of potato infusion, 20 g of dextrose) at room temperature for seven days. The fungal mycelium was washed in distilled water using a sieve to remove all the culture medium. Next, the sample was dried with a paper towel to remove moisture and placed in the freezer at -20 °C until the analysis.
The genomic DNA of the fungi was extracted using a DNA extraction kit, following the manufacturer's recommendations (Axygen Biosciences, USA). Morphotyping was carried out by analyzing the genetic variability of the ISSR (Inter Simple Sequence Repeat Amplification) and IRAP (Inter Retrotransposon Amplified Polymorphism) molecular markers.
The identification was performed through partial sequencing of the internal transcribed spacer (ITS) of the rDNA region of representatives of each morphotype group. The oligonucleotides ITS 4 (5'-TCC TCC GCT TAT TGA TAT GC-3') and ITS 5 (5'-GGA AGT AAA AGT CGT AAC AAG G-3') (White et al., 1990) were used to amplify the 18S to 28S intergenic region. The reaction mixture of 35 μL contained 22.35 μL of ultra-pure water, 3.5 μL of 10x PCR buffer, 1.5 μL of MgCL 2 , 2.8 μL of dNTP, 1 μL of ITS4 primer, 1 μL of ITS5 primer, and 0.3 μL of the Taq polymerase enzyme. The amplification conditions were performed in an AmpliTherm Thermal Cycler as follows: initial denaturation at 94 °C for 2 min; 35 cycles of 94 °C denaturation for 45 sec, annealing at 50 °C for 45 sec, and elongation at 72 °C for 1 min; and the final extension at 72 °C for 10 min.
The amplification products of the gene 16S (bacteria) and the ITS region (fungi) were purified according to Dun and Blattner (1987). To quantify the DNA, 1 μL of the product was loaded into a 0.8% agarose gel for gel electrophoresis. The sequencing was performed using the Sanger method with a Big Dye Kit on an ABI 3100 (Applied Biosystems). The sequences were compared in the GenBank database (NCBI, 2019) against known sequences through a search for similarity using BLASTN.

Microbiolization of the Butia purpurascens and Butia archeri seeds
The bacteria were selected according to the best results on the antibiosis test. To prepare the bacterial suspensions, the isolates were cultivated in a nutrient broth medium (3 g meat extract, 5 g peptone) separately for 24 hours.
After this period, the concentration of the inoculation was adjusted to OD 600 = 0.1 with a sterilized saline solution (0.85 NaCl).
The seeds were submerged in a GA 3 (200 mg L -1 ) solution for 48 hours in a germinator set to 30 °C and then submerged for 30 min in bacterial inoculants. For the Butia purpurascens seeds, the following microorganisms were used: control (nutrient broth), the rhizospheric bacteria Bacillus amyloliquefaciens BP1RB, B. amyloliquefaciens BP70RB, B. subtilis BP186RB, and B. methylotrophicus BP35EB (endophytic). For Butia archeri, seeds taken from the Bacillus sp. BA68EB (endophytic) and Brevibacillus brevis BA89R (rhizospheric) and the control were used.
Once removed from the inoculum, the seeds were placed on glass plates (120 × 150 mm) containing Germitest  paper to germinate. These were moistened with autoclaved distilled water in the amount of 2.5 times the weight of this paper and kept in a Mangelsdorf germinator set to 30 °C for 30 days.
The experiment used a completely randomized design with five repetitions each. The percentages of germination, contamination by fungi and bacteria, and hard seeds were evaluated. The averages were compared through the Tukey's test (5%).

Germination of Butia purpurascens seeds in different substrates
There was no difference between the cultivation for the B. purpurascens seeds in terms of the percentages of germination, hard seeds, and contamination by fungi (Table 1). A difference between the germination substrates was observed only in the percentage of contamination by bacteria: the glass plate with Germitest  paper yielded the least contamination by bacteria at only 0.8% (Table 1).
We highlight that the contamination occurred mainly near the lesions during seed extraction and scarification, which may have served as a gateway for saprophytic microorganisms (fungi and bacteria), and possibly reduced the germination rates of the species in these substrates.

Identification of the microorganisms
The grouping of the fungal isolates from seeds of the three evaluated cultivation substrates into morphotypes indicated the presence of 17 groups for B. purpurascens, totaling 67 fungi and, for B. archeri, 10 groups with 40 fungi. These were distinguished by macroscopic characteristics. After molecular identification of one representative of each group through the amplification profile of the molecular marker ISSR and/or IRAP of the different strains (data omitted), identical profiles could be detected, enabling taxonomic identification.
At the species level, an identity above 97% of the ITS region was used as a criterion of similarity (O'Brien et al., 2005).
The antagonistic bacteria selected and identified were exclusively of the genus Bacillus, and of the species amyloliquefaciens, brevis, methylotrophicus, and subtilis ( Table 4). Members of this genus are commonly found in environments with soil and water. Some of its species have antifungal activity against plant pathogens and are capable of inhibiting mycelial growth and the germination of conidia, and thus play important roles in the infection cycle of pathogenic agents (Song et al., 2014).
In the evaluation of the results for growth inhibition of the deteriorative fungus of B. archeri (A. niger BA163DF) seeds, no difference was found between the microorganisms with Bacillus sp. BA68EB and B. brevis BA89RB when compared to the control (Table 6). Table 1. Porcentages of germination, hard seeds, and contamination by fungi and bacteria in Butia purpurascens seeds in different substrates.

Microbiolization of Butia purpurascens and Butia archeri seeds
In the microbiolization of B. purpurascens and B. archeri seeds using endophytic and rhizospheric strains of Bacillus sp., no differences were found in the percentages of germination and contamination by fungi, the latter of which varied between 19.3 to 39.3% (Table 7). There was also no difference between the solutions for seed germination, varying from 24.3 to 41.0% for the yatay palm.
Contamination by fungi and bacteria was noted for B. archeri after one day of cultivation, mainly in the regions of lesions caused by the extraction and the scarification process. We observed that initially there was intense release of seed exudates, followed by fungal contamination. However, there was no difference in fungal and bacterial contamination between bacterial treatments, which varied between 60% and 63.3%.
A low percentage of germination was found, with a minimum of 16.6% and maximum of 22.5%. There was no difference in germination efficiency between the bacteria evaluated to control the contamination during the evaluated period (Figure 2).

Discussion
The results observed in this study for the percentage of germination of B. purpurascens seeds in different substrates was lower than that observed by Fior et al. (2011) with Butia capitata (Martius) Beccari seeds, which reached 90% germination when kept in Gerbox plastic boxes containing autoclaved sand for 150 days. The germination of wax palm (Copernicia alba Morong.) seeds was also favored: it reached 67% when sowed on paper at a temperature between 20 and 30 °C (Masetto et al., 2012). Rubio Neto et al. (2014) obtained 63.8% germination of Acrocomia aculeata (Jacq.) Lood. ex Mart. seeds when they were scarified and kept on rolls of Germitest  paper.
In this study, the seeds were not immersed in a gibberellic acid solution, which is an alternative to increase the germination percentages of seeds with physiological dormancy. Studies consider immersing the seeds in a gibberellic acid solution and removing the opercular tegument to be an alternative method that can promote increased germination in macaw palm seeds Rodrigues Junior et al., 2013).
For the B. purpurascens seeds, the bacterial isolate B. subtilis BP186RB showed the best capacity for suppressing the growth of Neodeightonia phoenicum BP91DF and Penicillium purpurogenum BP110DF, the two deteriorative fungi tested. However, for B. archeri seeds, no difference were observed between the microorganisms with Bacillus sp. BA68EB and B. brevis BA89RB compared to the control in inhibiting the growth of deteriorative fungi.
Bacillus sp. are capable of synthesizing a structurally diverse group of antimicrobial compounds of low molecular weight that suppress phytopathogens and cyclic peptide derivatives, especially lipopeptides (Kilimushi et al., 2017). Gond et al. (2015), through a MALDI-TOF-MS analysis, identified antifungal groups such as surfactin, iturin, and fengycin that exhibited strong antibacterial activity, consequently inhibiting the growth of filamentous fungi by antagonizing sterols, phospholipids, and oleic acid in fungal membranes, thereby protecting against pathogenic agents. Song et al. (2014) performed a study for controlling ginseng root rot, caused by the fungus Fusarium, through the use of B. amyloliquefaciens. The study showed, through scanning electron microscopy, that the hyphae of the pathogen were twisted and withered with bacterial treatment, which can be a symptom of direct damage by antifungal substances.
In the microbiolization of B. purpurascens and B. archeri seeds performed with microbiological solutions produced based on endophytic and rhizospheric strains of Bacillus sp., no differences were observed in the percentages of germination and contamination by fungi.
For B. archeri, there was contamination by fungi and bacteria after one day of cultivation, mainly in the regions with lesions caused by the extraction and scarification process.
This is the first study for B. purpurascens and B. archeri that shows major problems with contamination during germination, exhibiting low percentages of normal seedlings. However, further studies should be carried out to further validate the use of biological control in these species. Means followed by the same letter in the column are not significantly by Tukey test (P > 0.05).