Bioactivity of fungi isolated from coconut growing areas against Rhynchophorus palmarum

– The objective of this work was to evaluate the chitinolytic activity and bioactivity of fungi isolated from the palm weevil ( Rhynchophorus palmarum ) and from soil samples from coconut ( Cocos nucifera ) crops against the insect itself. Initially, to determine the chitinolytic properties of the isolated fungi, their ability to hydrolyze chitin in a liquid culture medium was evaluated. Then, preliminary pathogenicity assays were carried out, using the bean weevil ( Acanthoscelides obtectus ) as a reference, to select the fungal isolates to be used in the experiments with the palm weevil . Finally, the bioactivity of two selected entomopathogenic fungi on palm weevil larvae and adults was assessed. There was no direct correlation between chitinolytic activity and pathogenicity capacity on the bean weevil nor between the isolates and bioactivity on the palm weevil. Beauveria bassiana CSU9 shows the highest activity on palm weevil larvae and adults, with a median lethal time of 0.8 and 14.4 days, respectively.


Introduction
Coconut (Cocos nucifera L.) is one of the most popular crops around the world. In 2020, the coconut production worldwide almost reached 61 million tons, i.e., 12.3 million tons per hectare, with Indonesia, Philippines, India, Sri Lanka, and Brazil standing out as the largest producers, representing 31, 23, 18, 4.2, and 3.8%, respectively, of the total produced (Statista, 2020). Although Colombia only produces 0.145 million tons per year of coconut fruit -destined for domestic consumption and corresponding to 0.24% of the world market -, the country has a high potential for the growth of this crop due to its specific climatic conditions, which favor yields of 7.2 Mg ha -1 , above the world average of 5.0 Mg ha -1 (Statista, 2020). However, as in other countries, Colombian coconut production is affected by several phytosanitary problems, among which the palm weevil (Rhynchophorus palmarum L.) is one of the most important (León-Martínez et al., 2019).
The palm weevil causes mechanical damage to palms, especially in the early stages of plant development, and is also the main vector of the Bursaphelenchus cocophilus (Cobb) Baujard nematode, the causal agent of the red ring disease (RRD) (Calvache et al., 1995). This nematode has a wide range of hosts, affecting several crops such as oil palm (Elaeis guineensis Jacq.), peach palm (Bactris gasipaes Kunth), and date palm (Phoenix dactylifera L.). The RRD causes yellowing and the progressive death of older leaves and shortening of emerging leaves, which affect photosynthetic processes and lead to palm death (Kambiré et al., 2021). The presence of B. cocophilus has been reported in Central and South America, as well as in the Caribbean islands. In Colombia, the RRD and palm weevil have affected more than 70,000 planted hectares, reducing yield by more than 50% in the main coconut growing areas in Cauca, Nariño, and Antioquia (Chinchilla, 2010); annual losses have been estimated at USD $5.5 billion. Therefore, the RRD and palm weevil are considered the most threatening phytosanitary problems of palm crops (León-Martínez et al., 2019).
For the control of the palm weevil, insecticides based on cypermethrin, benzoylureas, and carbamates have been used (El-Bokl et al., 2010;El-Shafie et al., 2011). However, these types of insecticides have no effect on larval instars of palm weevil with cryptic habits. Therefore, biotechnological applications and biological control agents have emerged as potential alternatives (Mazza et al., 2014;Santana et al., 2019). In this scenario, the use of entomophatogenic fungi (EPF) has proven to be effective in pest control (Awan et al., 2021;Poveda, 2021); however, few known studies have been conducted to find natural enemies of the palm weevil. EPF can penetrate the cuticle (chitin) of the insects and release enzymes and toxins that reduce immune responses in hosts, preventing them from expressing resistance mechanisms (Hussain et al., 2016). In this case, the chitinolytic activity is a relevant factor for the selection of fungi with a high infectious capacity and pathogenic potential.
The objective of this work was to evaluate the chitinolytic activity and bioactivity of fungi isolated from the palm weevil and from soil samples from coconut crops against the insect itself.
Ten grams of homogenized soil were mixed to 90 mL saline solution (0.85% NaCl) and diluted at a ratio of 1:10. Then, 100 µL of the 10 -2 and 10 -3 dilutions were added to sterile Petri dishes containing 15 mL potato dextrose agar (PDA, 39 g L -1 ) supplemented with antibiotics specific for bacteria, i.e., with 3.4 mg L -1 ampicillin and 0.5 g L -1 streptomycin. The plates were incubated at 26°C, in the dark, for at least four days. The fungi isolates were purified and re-cultured in fresh PDA under the same conditions to obtain monosporic cultures (El Kichaoui et al., 2017). Conidia were mixed in 10 mL of 0.1% Tween 80 in sterile distilled water, and the suspensions were diluted and adjusted to 50-100 spores per millimeter. A volume of 100 µL conidia suspension was added to sterile Petri dishes containing 15 mL PDA and incubated at 26°C, in the dark, for at least four days.
Thirty palm weevils naturally infected were washed with 1.0% hand soap and placed in Petri dishes containing a moistened piece of cotton. After ten days, a portion of the mycelium developed on the insects was added to 10 mL saline solution (0.85% NaCl) and vortexed for 2 min. A final suspension of 1.0 mL was used to isolate the fungi following the procedure previously described. Then, 20 palm weevils were placed individually in plastic boxes (30×16×8.0 cm), whose lids had several perforations. Sugarcane slices (diameter of 3.0 cm and length of 7.0 cm) placed on wet absorbent paper were used to feed the insects and as an oviposition site for the females. The sugarcane and wet absorbent paper were renewed once a week, and emergent larvae were collected weekly.
The palm weevil larvae for the bioactivity assays were placed individually in 21 mL cups and fed with red apple slices. Afterwards, the palm weevil larvae that would be used for rearing were placed individually in 21 mL cups with 5.0 g meridic diet, containing 18.5 agar, 50 g L -1 Levapan bread yeast, 45 g L -1 wheat flour, 45 g L -1 corn flour, 1.6 g L -1 sorbic acid, 4.0 g L -1 ascorbic acid, 3.1 g L -1 Pharmaton dietary supplement, and 0.5 mg L -1 tetracycline (El-Shafie et al., 2013). Recipients and food were renewed weekly. After 60 days, the larvae were transferred to 50 mL falcon tubes containing coconut fiber to favor pupal development. The conditions established for all procedures were: photoperiod of 12 hours dark/light, relative humidity (RH) of 80%, and temperature of 30°C.
The chitinolytic activity was determined by the dinitro salicylic acid (DNS) method (Sandhya et al., 2004). Aliquots of 0.5 mL cell-free supernatant were added to the test tubes containing 0.5 mL of 0.5% colloidal chitin in phosphate buffer (pH 5.5) and 1.0 mL sterile distilled water. The used controls were: substrate, with 0.5 mL of 0.5% colloidal chitin in phosphate buffer and 1.5 mL sterile distilled water; and enzyme, with 0.5 mL cell-free supernatant and 1.5 sterile distilled water. Absorbance was measured in aliquots of 200 µL, at 540 nm, in the 680XR microplate reader (Bio-Rad, Hercules, CA, USA). One unit (1 U) of chitinolytic activity was defined as the amount of enzyme required to release 1.0 μg reducing sugars per millimeter per minute from a 0.5% colloidal chitin solution under the reaction conditions (Sandhya et al., 2004).
Conidial suspensions were prepared from fungal isolates grown on PDA, at 26°C, for ten days. The conidia were collected and placed in sterile falcon tubes containing 20 mL of 0.05% Tween 80 (v/v) in sterile distilled water. Spore concentration was adjusted to 10 8 mL -1 conidia. The viability of the conidia and the purity of the suspension were estimated by the percentage of germination of the conidia in PDA. The bean weevil [Acanthoscelides obtectus (Say)] was used in preliminary pathogenicity assays to select the fungal isolates that would be used in the experiments with the palm weevil. The bean weevil has the typical body structures of coleopterans and a short life cycle, besides being easily multiplied (Silva, 2017); in the present study, the bean weevil was reared from insects collected from commercial white bean (Phaseolus vulgaris L.) stocks. The insects were immersed in 0.5% sodium Pesq. agropec. bras., Brasília, v.57, e02882, 2022 DOI: 10.1590/S1678-3921.pab2022.v57.02882 hypochlorite for 10 s, and five insects were placed inside sterile glass jars containing 5 U of beans and then sprayed with 700 µL (10 8 mL -1 conidia) of conidial suspension. The control was sprayed with 700 µL of 0.01% Tween 80. The sterile glass jars were covered with a mesh cloth and incubated for 16 days at 12 hours dark/light photoperiod, 75% RH, and 29°C. Mortality was recorded every day to determine median lethal time (LT 50 ). For this, a nonlinear probability model was obtained with the options of discrete choice models (Probit) and a completely random statistical model was analyzed by the method of repeated measures over time (α=0.05). The treatments consisted of the fungal isolates that showed the highest chitinolytic activity by the DNS method, evaluated in triplicate (Rodríguez-González et al., 2018b). The species Beauveria bassiana CSU9 was included in the present study since it has been widely reported as an EPF against coleopteran insects (Rodríguez-González et al., 2018a). Afterwards, palm weevil larvae were immersed in 0.5% sodium hypochlorite for 10 s, dried, and dipped in the conidia suspension for 20 s. The inoculated larvae were individually placed on a moistened and sterilized napkin inside plastic boxes (2.0×6.0×3.0 cm) containing a cotton swab impregnated with a sterile 10% sucrose solution. The plastic boxes were kept for 30 days at a 12 hour dark/light photoperiod, 30°C, and 90% RH. Larval mortality was recorded every two days, and the LT 50 value was determined. Two of the fungal isolates that showed the lowest LT 50 values on the bean weevil were selected and used as treatments, which were evaluated on three different dates, with three replicates of five individuals each. For the control, five larvae were dipped in 0.01% Tween 80 only. Finally, palm weevil adults -captured and quarantined in the laboratory under a 12 hour dark/light photoperiod, 29°C, and 90% RH for 15 days -were disinfected by immersion in 0.05% sodium hypochlorite for 30 s, rinsed three times with sterile water, dried, and dipped in the conidia suspension for 20 s. The insects were individually placed in plastic boxes (16×12×7.0 cm) containing portions of sugarcane as food, and the LT 50 values were estimated. In all experiments, to verify if they had been inoculated in dead larvae or insects, the fungi were re-isolated in PDA.
The initial identification of the isolates was done by microscopic observation using morphological keys (Seifert et al., 2011). The fungal isolates selected through the preliminary pathogenicity tests were identified. Their DNA was quantified using the NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). Polymerase chain reactions (PCRs) were performed in the iCycler (Bio-Rad, Hercules, CA, USA) in a 30 µL reaction mix containing 1.0 X PCR buffer, 1.5 mmoL -1 MgCl2, 0.2 mmoL -1 dNTP, 1 U Taq DNA polymerase, 50-100 µg L -1 template DNA, and 0.2 μmoL -1 of each ITS1/ ITS4 or EF1/EF2 primer. All PCRs were performed for 30 cycles, which consisted of denaturation for 1 min at 94°C, hybridization for 1 min at 50°C, extension for 1 min at 72°C, and final extension for 5 min at 72°C. The amplified products were analyzed by 1.0% (w/v) agarose gel electrophoresis and visualized with the GelRed Nucleic Acid Gel Stain (Biotium: Growing Products for Science, Freemont, CA, USA). The purified PCR amplicons were further sequenced by Macrogen (Seoul, Korea) and compared with those reported by BLAST (National Library of Medicine, Bethesda, MD, USA) to confirm the genetic identity of the isolates.
The analysis of variance (ANOVA), at a significance level of 5%, and multiple comparison tests were performed to determine the differences among treatments. Tests for normality of residuals and homogeneity of variances were performed before the ANOVA. All statistical analyzes were carried out with the Statgraphics Centurion XVl software (Statgraphics Technologies, Inc., Plains, VA, USA).

Results and Discussion
A total of 33 fungi belonging to five genera were isolated: 17 of Trichoderma spp., 12 of Purpureocillium spp., 2 of Paecilomyces spp., 1 of Beauveria sp., and 1 of Metacordyceps sp. (Table 1). Of these, based on the morphological differences between isolates of the same genera, 15 were selected to be used in the chitinolytic activity assays. The Trichoderma harzianum T12 isolate was also included in the present study as a reference since it has been reported as a chitinase-producing fungus with an enzymatic activity greater than 9 U (Sandhya et al., 2004;Abu-Tahon & Isaac, 2020). By comparing the sequences of the selected isolates with those of the isolates stored in the GenBank database (National Library of Medicine, Bethesda, MD, USA), it was possible to confirm the At 48 hours, the highest values for the chitinolytic activities were: 24.9±5.9 U, with M. chlamydosporia CSU10; 7.6±1.1 U, with Purpureocillium sp. CSC1; 7.5±0.2 U, with Trichoderma sp. CU1; 6.5±0.7 U, with T. virens CSC11; and 6.3±0.1 U, with Purpureocillium sp. CSU2 (Table 1). These values were 4.0, 1.2, 1.2, 1.0, and 1.0 times, respectively, that obtained with T. harzianum T12, which was of 6.3±0.1 U. Consequently, the four listed isolates were considered to have great potential for insect control and were used in the pathogenicity assays with the bean weevil.
The lowest LT 50 values for the bean weevil were: 3.2±0.8 days, obtained with T. virens CSC11; 5.1±1.0 days, with B. bassiana CSU9; 5.8±0.8 days, with Purpureocillium sp. CSU2; and 6.1±1.0 days, with Purpureocillium sp. CSC1. Both B. bassiana CSU9 and T. virens CSC11 showed LT 50 values 51 and 69% lower than those of the control, respectively (Figure 1). Surprisingly, M. chlamydosporia CSU10 and Trichoderma sp. CU1 caused the lowest pathogenicity on the bean weevil despite exhibiting the highest chitinolytic activities (Table 1). This could be explained by the fact that the pathogenicity of the EPF is not only defined by its ability to penetrate the insect cuticle but also by other factors such as its capacity to release different enzymes and toxins necessary to weaken the insect's immune system (Mondal et al., 2016;Boguś et al., 2017). In addition, although Silva (2017) concluded that the cuticle of adult coleopteran insects is mainly composed of chitin, Mondal et al. (2016) found that the external cuticle of palm weevil larvae is composed of other major compounds, as lipids and proteins, which was not considered in the present study. Both T. virens CSC11 and B. bassiana CSU9 were selected for further experiments since the virulence of Trichoderma sp. and B. bassiana isolates has been widely reported against insects, including beetles (Ricaño et al., 2013;Berini et al., 2016;Rodríguez-González et al., 2018a).
Beauveria bassiana CSU9 showed a better entomopathogenic activity than T. virens CSC11 in the palm weevil larvae, with LT 50 values of 0.8 and 29.6 days, respectively ( Figure 2). However, in adults palm weevils, there were no differences between the LT 50 of B. bassiana CSU9 and T. virens CSC11, with 14.4 and 16.7 days, respectively. Considering the development of the mycelium in the legs, spittle, and elytra of the treated insects, B. bassiana CSU9 proved to be effective against palm weevil larvae and adults, with a respective LT 50 of 0.8 and 14.4 days. The mode of action of B. bassiana involves cuticle penetration through the integument due to the carbonand nitrogen-rich resources present in the insect bodies (Islam et al., 2021). Moreover, this species has the ability to evade the immune response of insects (Lovera et al., 2020), a process that is mediated by the physical pressure exerted by the hyphae on the cuticle and by the enzymatic activity on its components. Xiao et al. (2012) observed that B. bassiana has an advanced adhesion mechanism involving hydrophobins that mediate cell surface hydrophobicity and virulence. Furthermore, two nonribosomal peptide synthetase genes encoding bbBeas and bbBsls, responsible for the synthesis of the insecticidal virulence factors beauvericin and bassianolide, respectively, have been identified recently (Batool et al., 2020); both may be effective strategies to facilitate pathogenic interactions with insects. Trichoderma sp., however, presents mycoparasitism as a mechanism of action, produces insecticidal secondary metabolites, and competes with host insects through more complex and slow processes (Poveda, 2021). León-Marinez et al. (2019) collected isolates of B. bassiana from a commercial oil palm plantation and reported mortalities of palm weevil adults from 36 to 47% at 21 days after inoculation (LT 50 > 21 days). Dembilio et al. (2012) used concentrations of B. bassiana higher than 10 9 mL -1 conidia to increase the mortality of Rhynchophorus ferrugineus (Olivier, 1790) adults after 21 days of treatment (LT 50 > 21 days). Compared with the adult insects of these other species, T. virens CSC11 showed lower LT 50 values (< 21 days) in the present study.