Host-associated bioprospecting from Opisina arenosella: a Beauveria bassiana isolate against red palm weevil

Abstract

Host-associated bioprospecting of naturally infected insects provides a valuable approach for recovering indigenous entomopathogenic microorganisms adapted to local environments. In this exploratory study, seven microbial isolates (Four fungal and three bacterial) were recovered from naturally infected Opisina arenosella exhibiting signs of mycosis. Their pathogenicity was evaluated under laboratory conditions against larvae of the red palm weevil (RPW, Rhynchophorus ferrugineus). Among the recovered isolates, a single fungal isolate was identified as Beauveria bassiana (Zeb-01) and showed greater virulence compared with the other isolates obtained in this study. Morphological characterization using light and scanning electron microscopy, combined with molecular identification based on ITS and EF1-α sequencing, confirmed the identity of Zeb-01 and described its hyphal and conidial features. In laboratory bioassays, Zeb-01 caused dose- and time-dependent mortality, reaching 85% larval and 86% adult mortality at 1.0 × 10¹⁰ conidia/mL within 10 days. At the lowest tested concentration (1.0 × 10⁶ conidia/mL), mortality rates of 35% and 50% were recorded for larvae and adults, respectively. When evaluated under the same experimental conditions alongside a commercial B. bassiana formulation, Zeb-01 exhibited lower LC₅₀ and LT₅₀ values. These results document the isolation, identification, and laboratory evaluation of a single native B. bassiana isolate recovered from O. arenosella, and demonstrate its entomopathogenic activity against RPW, indicating its potential relevance for integrated pest management programs (IPM).

Graphical abstract

Introduction

The red palm weevil (RPW), Rhynchophorus ferrugineus (Olivier) (Coleoptera: Curculionidae), is a globally invasive pest that inflicts severe economic damage on key palm species, including coconut (Cocos nucifera L.), date (Phoenix dactylifera L.), and oil palm (Elaeis guineensis Jacq.) (Liu et al. 2024; Nurashikin-Khairuddin et al. 2022; Sudalaimuthuasari et al. 2024). Originally endemic to South Asia, its distribution now extends the Middle East, Africa, the Mediterranean, and extensive regions of China, threatening palm production across at least 15 provinces (Abdel-Banat and El-Shafie 2023; Giblin-Davis et al. 2013; Li et al. 2009). The cryptic feeding activity of both larvae and adults, who tunnel through the meristem and vascular tissues, compromises the palm’s structural integrity, frequently causing catastrophic collapse (Aziz 2024; Tagliavia et al. 2014). Current management relies heavily on a combination of phytosanitary measures, pheromone-baited trapping, and broad-spectrum insecticide applications (Milosavljević et al. 2022; Rehman and Mamoon-ur-Rashid 2022). However, these strategies are increasingly inadequate: insecticides promote resistance, induce significant non-target effects, and raise environmental concerns; trapping is only effective against adults; and the removal of infected palms is economically unsustainable (Ali 2023; El-Shafie and Faleiro 2017; Wakil et al. 2018). Consequently, there is a critical unmet need to develop sustainable, cost-effective, and stage-specific control tactics to bolster integrated pest management (IPM) frameworks for this pervasive pest.

Isolation of indigenous microbial agents from naturally infected insect cadavers provides a valuable approach for recovering entomopathogenic microorganisms adapted to local ecosystems, which may serve as potential biocontrol candidates. Among these microbial isolates, entomopathogenic fungi (EPF) are particularly relevant due to their natural pathogenicity toward arthropods, broad host ranges, and ability to trigger epizootics (Bamisile et al. 2021; Dara et al. 2019; Gupta et al. 2023; Idrees et al. 2022). EPF infect hosts through direct cuticular penetration, producing enzymes and toxins that suppress host immunity, ultimately causing systemic mycosis and mortality (Hong et al. 2024). Importantly, EPF are also considered compatible with integrated IPM strategies and are associated with a lower risk of resistance development. Among EPF, Beauveria bassiana (Bals.-Criv.) Vuill. is one of the most widely studied species, with demonstrated efficacy against coleopteran, lepidopteran, and hemipteran pests (Irsad et al. 2023). Nevertheless, the success of field applications is strongly influenced by strain-specific traits, such as virulence, competitive fitness, and adaptability to local environmental conditions, underscoring the relevance of isolating locally adapted strains preconditioned to the prevailing climatic variables, host interactions, and ecological dynamics of the target region.

Recent studies have shown that entomopathogens recovered from ecologically related or sympatric insect hosts can exhibit cross-pathogenicity, broadening their potential use in pest suppression (Abrar et al. 2023; Sutanto et al. 2021). This approach is particularly relevant for palm agroecosystems, where multiple insect pests often co-occur. In this context, the coconut black-headed caterpillar, Opisina arenosella Walker (Lepidoptera: Oecophoridae), a major defoliator of palms in tropical Asia, represents a practical host for isolating entomopathogenic strains. As O. arenosella and the RPW coexist in coconut agroecosystems and frequently cause damage individually or synergistically, recovering entomopathogens from O. arenosella offers a strategy for identifying potential biocontrol agents against RPW. Although previous studies have confirmed the pathogenicity of B. bassiana against RPW (Alwaneen et al. 2024; Irshaid et al. 2024; Ment et al. 2023; Zhong et al. 2025), most have focused on adult susceptibility, leaving gaps in understanding larval responses and comparative strain performance. Systematic evaluation of isolates obtained from alternate palm-associated pests may help identify strains with differential virulence profiles and host interactions, contributing to IPM strategies.

To address these research gaps, microbial isolates were recovered from cadavers of O. arenosella collected in coconut plantations and screened for pathogenicity. Among the recovered isolates, a single fungal isolate, identified as B. bassiana and designated Zeb-01, exhibited comparatively higher virulence under laboratory conditions. Its morphological features were examined using optical and scanning electron microscopy (SEM), and its identity was confirmed through internal transcribed spacer (ITS) and EF1-α gene sequencing. The virulence of Zeb-01 was subsequently evaluated against both larval and adult stages of the RPW under controlled laboratory conditions.

Materials and methods

Insect sources for the test

RPW were reared under controlled laboratory conditions at the Biological Control Laboratory of the Coconut Research Institute, Chinese Academy of Tropical Agricultural Sciences (Lv et al. 2025). To establish a culture, an artificial diet was formulated by thoroughly mixing ground coconut tissue and wheat flour at a ratio of 9:1 (w/w). Distilled water was gradually added while mixing until a homogenous, moist substrate was achieved that was sufficiently firm to support larval burrowing and development. For oviposition, 20 healthy adults of RPW (10 males and 10 females) were introduced into a plastic rearing box measuring 30 × 20 × 10 cm (Length × Width × Height). Approximately 2 kg of the prepared artificial diet was evenly spread at the base of the box. Fresh coconut husks were placed on top of the diet to mimic a natural oviposition site and stimulate egg-laying behavior. The rearing box was sealed with a perforated plastic lid containing multiple small holes to ensure adequate ventilation while preventing adult escape. The rearing conditions were maintained in an environmental chamber at 25 ± 1 °C, relative humidity ≥ 70%, and a photoperiod of 16:8 h (Light: dark). The RPW adults were monitored daily, and oviposition typically occurred within 3–5 days. After two weeks, the larvae were transferred into the same insect rearing boxes containing fresh larval diet, and their growth and development were monitored. The diet was replaced approximately every 7 days, with the replacement frequency flexibly adjusted according to the actual feeding status of the larvae. Subsequently, only vigorous, healthy third-instar larvae (as they are highly susceptible to the test agents and constitute the most damaging instar in field populations due to their feeding activity) and active adults were selected for subsequent bioassays.

Isolation and culture of microbial pathogens

Two distinct methods were employed for the isolation of microbial pathogens from infected Coconut Black-headed Caterpillar (O. arenosella) cadavers collected from the field. In the first method, fungal spores visibly growing on the surface of the cadavers were carefully scraped using a sterilized inoculation loop under aseptic conditions. These spores were then directly transferred and streaked onto Potato Dextrose Agar (PDA; 200 g potato, 20 g glucose, 15 g agar powder, distilled water to 1,000 mL, sterilized at 121℃ for 15 min containing 0.1 g/L chloramphenicol) plates for streak culture. After single colonies emerged, those with distinct morphologies were picked and streaked onto fresh PDA plates for further purification. This process was repeated several times until pure culture isolates were obtained. The second method was the tissue isolation method (Fang 1998), where the cadavers were immersed in 70% ethanol for 1 min and rinsed with sterile water. Next, surface sterilization was performed using a 0.1% mercuric chloride solution, and washed three times with sterile distilled water. Finally, the surface moisture of the cadavers was carefully absorbed using sterile filter paper to avoid interference during the subsequent inoculation. The cadavers were then cut into several segments with sterilized scissors and inoculated onto PDA medium. The inoculated tissues were placed on separate sterile petri dishes, sealed with Para film, and incubated at 25 ± 1℃ with a relative humidity of 75 ± 5% and photoperiod of 14 L:10 D. Once new hyphae emerged on the surface of the cadavers, which indicated fungal growth, the edge of the hyphae was carefully transferred to fresh PDA medium for streak culture to achieve purification. This process was repeated until pure fungal strains were obtained. The purified strains were then inoculated onto PDA medium and cultured under the previously described environmental conditions for 15 days. During this period, the morphology of the colonies was closely observed. The isolated strains were preliminarily identified based on morphological characteristics.

Primary screening for virulence of microbial pathogens against RPW larvae

The seven obtained isolates were separately diluted with sterile water containing 0.05% Tween-80 to concentrations ranging from 1.0 × 10¹⁰ to 1.0 × 10⁸ conidia/mL. The virulence of each isolate against larvae of RPW was determined using the larval immersion method. Third-instar larvae of uniform size and high activity were selected for the experiment. The larvae were immersed in the prepared spore suspensions of different concentrations for 1 min, then transferred to the aforementioned rearing boxes containing larval diet, and incubated at 25 ± 1 ℃, 75 ± 5% relative humidity (RH), under an 8:16 h light-dark photoperiod. Larval mortality was examined and recorded 10 days after treatment. Each concentration was applied to 20 larvae, with three replicates. Sterile water supplemented with 0.05% Tween-80 was used as the control.

Morphological identification and biological characterization of the strain

Based on the above bioassay results, the subsequent experiments mainly focused on the in-depth study of Isolate-1, which exhibited the highest pathogenicity. Purified fungal strains were cultured on PDA and incubated for 15 days under previously described conditions, with colony morphology monitored throughout. Mycelia, conidia, and conidiogenous structures were examined using light microscopy and SEM, and conidial dimensions were measured from at least 20 spores. Five representative strains were selected for SEM observation in this study to characterize the ultra-morphology of conidia and mycelia. Preliminary identification was based on morphological characteristics consistent with Beauveria spp (Rehner et al. 2011), including colony color, texture, conidial morphology and size, among other traits. Mycelial plugs (5 mm) were transferred to fresh PDA plates, and colony diameters were measured daily for 15 days using the cross-intersection method (three replicates per strain) to calculate mean mycelial growth rates (mm/day). After incubation, spores were harvested with sterile 0.05% Tween-80, filtered through sterile gauze, and quantified using a hemocytometer. All assays were performed in triplicate.

Molecular biological identification of the pathogenic isolates

Genomic DNA of all pathogenic isolates was extracted using a standard fungal or bacterial genomic DNA extraction protocol. The DNA used for PCR had a concentration of 25 ng/µL; negative controls (sterile ultrapure water replacing the DNA template) were set up throughout PCR to eliminate nucleic acid contamination. The dissolved DNA was used as a template to amplify the ITS and EF1-α gene sequences of fungal isolates using universal primers. ITS region: ITS1 (5′- TCCGTAGGTGAACCTGCGG-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′) (Gebremariam et al. 2021). EF1-α gene: 983 F (5′-GCYCCYGGHCAYCGTGAYTTYAT-3′) and 1567R (5′-ACHGTRCCRATACCACCSATCTT-3′) (Carrillo-Benítez et al. 2013). For bacterial isolates, the 16 S rRNA was amplified with universal primers 27 F (5’-AGAGTTTGATCCTGGCTCAG-3’)/1492R (5’-GGTTACCTTGTTACGACTT-3’). PCR reactions (50 µL) contained 25 µL 2×Taq Master Mix, 1 µL of each 20 µM primer, 2 µL template DNA, and 21 µL ddH₂O. Amplification was performed at 94 °C for 4 min; 35 cycles of 94 °C for 1 min, 55 °C for 40 s, and 72 °C for 90 s; followed by a final extension at 72 °C for 10 min. PCR products were verified by 1% agarose gel electrophoresis and sequenced commercially (Sangon Biotech, Shanghai, China). Sequences were analyzed using NCBI BLAST, aligned with related sequences in MEGA 6.0, and a neighbor-joining phylogenetic tree was constructed. Tree topology and stability were assessed by bootstrap analysis with 1000 replicates.

Virulence assay of B. bassiana against adults and larvae of RPW

A purified B. bassiana strain was used for virulence assays. Mycelia and conidia were suspended in sterile 0.05% Tween-80, filtered through sterile defatted cotton to remove mycelial fragments, and a djusted to five concentrations (1.0 × 10¹⁰ to 1.0 × 10⁶ conidia/mL). Commercial B. bassiana (10 billion conidia per gram, Shanxi Lvhai Pesticide Technologies Co., Ltd. China) was prepared at three concentrations (1.0 × 10¹⁰ to 1.0 × 10⁸ conidia/mL) as a reference. Third-instar larvae and adults of RPW of uniform size and vigor were selected for the bioassay. The test insects were starved for 24 h prior to the experiment, then individually immersed in conidial suspensions for 1 min and transferred to individual rearing cups containing artificial diet (larvae: 10 g/cup; adults: 3 × 3 cm coconut husk) (District et al. 2024). Insects were maintained at 25 ± 1 °C, 75 ± 5% RH, with an 8:16 h light-dark cycle. Mortality was recorded every 24 h for 10 days, with fresh feed added as needed for the test insects, and insect death confirmed by lack of response to gentle probing and mycosis verified by white fungal growth on cadavers. Each treatment included three replicates of 20 larvae and five replicates of 10 adults; a blank control was treated with sterile 0.05% Tween-80. Mortality, corrected mortality (Abbott 1925), median lethal concentration (LC₅₀), and median lethal time (LT₅₀) were calculated.

Data analysis

Experimental data were organized in Microsoft Excel 2021 and analyzed using SPSS v21.0. Prior to conducting ANOVA, normality andhomogeneity of variance tests were performed on the data to ensure the fulfillment of analysis prerequisites. One-way ANOVA (P < 0.05)followed by Tukey’s post-hoc test was used to assess significant differences in RPW mortality across spore concentrations. To calculate theLT₅₀, LC₅₀, correlation coefficient (r), and corresponding 95% confidence intervals, Probit regression was performed using the dedicatedmodule in SPSS, based on the relationship between log-transformed fungal spore concentrations and probit-transformed mortality rates.

Results

Isolation, screening, and identification of microbial isolates

Seven microbial isolates were obtained from cadavers of naturally infected O. arenosella exhibiting typical mycosis symptoms using the tissue isolation method. These comprised four fungal isolates (Isolates 1–4) and three bacterial isolates (Isolates 5–7), as illustrated in Fig. 1. All isolates were individually evaluated for pathogenicity against RPW larvae under laboratory conditions at three concentrations (1.0 × 10⁸, 1.0 × 10⁹, and 1.0 × 10¹⁰ conidia/mL⁻¹) ), as summarized in Table 1. Significant differences in virulence were observed among the tested isolates (P < 0.05). Isolate-1 consistently exhibited the highest pathogenicity across all concentrations, causing 58.33 ± 1.67%, 63.33 ± 4.41%, and 85.00 ± 7.64% larval mortality at 1.0 × 10⁸, 1.0 × 10⁹, and 1.0 × 10¹⁰ conidia/mL⁻¹, respectively. In contrast, the remaining fungal isolates (Isolates 2 and 3) induced significantly lower mortality, not exceeding 33.33% even at the highest concentration, while Isolate-4 showed no detectable pathogenic activity. Among the bacterial isolates, Isolate-5 caused moderate mortality (maximum 25.56 ± 1.11%), Isolate-6 exhibited weak pathogenicity (10.00 ± 1.92%), and Isolate-7 had no measurable effect. Statistical analysis using Tukey’s test confirmed that Isolate-1 was significantly more virulent than all other isolates at each tested concentration. The colony color, texture, pigmentation, mycelial characteristics, conidiogenous structures, conidial morphology, and other phenotypic characteristics of all pathogenic isolates were observed. Molecular identification was performed using ITS and EF1-α gene sequencing for the fungal isolates and 16 S rRNA sequencing for the bacterial isolates, followed by BLAST comparison against reference databases. This analysis confirmed Isolate-1 as B. bassiana, Isolates 2 and 3 as Aspergillus, and the pathogenic bacterial isolates as Serratia marcescens. Among these, the B. bassiana isolate (designated Zeb-01) exhibited significantly higher virulence in preliminary screening assays. Consequently, Zeb-01 was selected for in-depth characterization and comparative evaluation of its efficacy against RPW.

Fig. 1.

Microbial isolates obtained from cadavers of the coconut black-headed caterpillar (O. arenosella)

Table 1.

Mortality rate (%) of isolated fungal and bacterial strains from naturally infected coconut black-headed caterpillars (O. arenosella Walker), against RPW larvae after 10 days of treatment

Isolates		Concentration (Conidia/mL)
		1.0 × 108	1.0 × 109	1.0 × 1010
		Mortality rate(%)
Fungi	Isolate-1	58.33 ± 1.67a	63.33 ± 4.41a	85.00 ± 7.64a
	Isolate-2	13.33 ± 1.92b	21.11 ± 2.94b	33.33 ± 3.85b
	Isolate-3	12.22 ± 1.11b	18.89 ± 2.22b	30.00 ± 1.92b
	Isolate-4	00.00 ± 0.00d	00.00 ± 0.00c	00.00 ± 0.00d
Bacteria	Isolate-5	11.11 ± 1.11bc	18.89 ± 1.92b	25.56 ± 1.11b
	Isolate-6	05.56 ± 2.22cd	07.78 ± 1.11c	10.00 ± 1.92c
	Isolate-7	00.00 ± 0.00d	00.00 ± 0.00c	00.00 ± 0.00d

Data are mean ± SE. Different letters within the same column indicate significant differences as determined by the Tukey’s test (P < 0.05).

Morphological observations and biological characterization of B. bassiana strain

The purified EPF strain of the most virulent fungal isolate-1 was cultured on PDA at 25 °C for 10 days for morphological analysis. Macroscopically, colonies were circular with regular margins, presenting a white cottony obverse surface with concentric zonation, and a reverse side with a central yellow region transitioning to a milky-white periphery with radial striations (Fig. 2a, b). Microscopically, the strain exhibited typical hyphomycete structures: optical microscopy showed hyaline, septate, and branched hyphae, and conidiophores bearing spherical, transparent conidia in tandem clusters (Fig. 2c, d). SEM further confirmed flask-shaped conidiogenous cells producing conidia singly or in small clusters; the conidia were globose to ovoid with a rough surface and a diameter of 1.43–2.36 μm in diameter (Fig. 2E and F). Based on these macroscopic and microscopic characteristics, which align with established descriptions, the strain was identified as B. bassiana. The strain had a mycelial growth rate of 4.06 mm/day, and achieved a sporulation yield of 2.32 × 10⁷ conidia/mL after 15 days of culture.

Fig. 2.

Colony morphology of the isolated fungal strain cultured on potato dextrose agar (PDA) for 10 days: a front (obverse) side, b back (reverse) side of the colony, c optical microscope images of sporulation structure and conidial pustules, d optical microscope images of hyphae and conidia, e SEM images of hyphae and conidiogenous cell, f SEM images of conidia

Molecular identification of B. bassiana strain

Molecular identification was conducted to confirm the taxonomic status of the isolate-1. The ITS region and EF1-α gene were amplified and sequenced, yielding 587 bp and 508 bp fragments, respectively. BLAST analysis of the ITS sequence revealed 100% identity with B. bassiana (OK331343), while the EF1-α sequence showed > 99% identity exclusively with B. bassiana. The ITS and EF1-α sequences have been deposited in the NCBI GenBank under accession numbers PQ821437 and PX641484, respectively. Phylogenetic trees were constructed using the neighbor-joining method in MEGA v6.0 to further validate the identity (Fig. 3a, b).

Fig. 3.

Phylogenetic tree constructed using the Neighbor-Joining method based on ITS region (a) and EF1-α gene (b) sequences, illustrating the relationship between the isolated strain and reference strains of B. bassiana and related species. Accession numbers are shown in parentheses; bootstrap values indicate branch support. The scale bar indicates the genetic variability in the genome

Based on both ITS (Fig. 3a) and EF1-α (Fig. 3b) sequences, the isolate clustered robustly within a clade of B. bassiana reference strains, with bootstrap support values of 89% and 98%, respectively. Metarhizium anisopliae was used as an outgroup. Integrating the molecular evidence with the corresponding morphological characteristics, the isolate was conclusively identified as B. bassiana and designated as B. bassiana strain Zeb-01. The strain was deposited in the China General Microbiological Culture Collection Center (CGMCC) in October 2025 under the accession number CGMCC 3.29439.

Morphological pathogenic effects of Zeb-01 spores on RPW larvae and adults

Spore suspensions prepared at various concentrations exhibited strong pathogenic effects against both larval (Fig. 4a, b) and adult (Fig. 4c, d) stages of RPW. Following inoculation, a progressive decline in locomotor activity was observed within 24–72 h, characterized by sluggish movement, reduced feeding, and eventual paralysis. Subsequently, the affected larvae and adults succumbed to death, with their bodies exhibiting pronounced signs of rigor mortis. After incubating the dead cadavers under moist conditions, fungal sporulation was detected on all cadavers. Upon further incubation of the cadavers under high-humidity conditions, early signs of fungal colonization became apparent within 24–48 h. White, filamentous hyphae initially emerged from natural body openings and mechanically weakened areas, such as the intersegmental membranes, the depressed regions of the cuticle, the base of the adult rostrum, and the tarsal joints at the base of the legs. As infection progressed, the hyphal network expanded radially across the insect surface. By 96–120 h post-inoculation, the entire cadaver was fully overgrown with a dense, woolly mat of hyphae, followed by abundant sporulation, resulting in a conspicuous white fungal covering. This external colonization reflects successful fungal invasion, proliferation, and sporulation, critical for secondary transmission in natural environments.

Fig. 4.

Lethal phenotype of isolated strain against RPW larvae (a and b) and adult (c and d)

Pathogenicity of the native B. bassiana isolate Zeb-01 against larval and adult RPW compared with a commercial formulation

The mortality rates of both larvae (F_(8,18) = 39.10, P < 0.001) and adults (F_(8,36) = 36.99, P < 0.001) of the RPW demonstrated a positive correlation with the increasing concentration of the strain’s conidial suspension. As shown in Table 2, the mortality rate of the control group in this experiment did not exceed 5%, thus no Abbott’s correction was required. At the highest concentration (1.0 × 10¹⁰ conidia/mL), Zeb-01 caused > 85.00% cumulative mortality in both life stages, outperforming the commercial B. bassiana strain, which induced 51.11% and 80.00% mortality in larvae and adults, respectively. Notably, even at the lowest concentration (1.0 × 10⁶ conidia/mL), Zeb-01 elicited substantial mortality (35.00% larvae, 50.00% adults), exceeding the commercial strain at higher concentrations (1.0 × 10⁸ conidia/mL; 20.00% larvae, 16.67% adults). These results indicate that Zeb-01 is highly virulent against RPW compared with a commercial formulation.

Table 2.

Pathogenicity of the isolate strain Zeb-01 and commercial B. bassiana against RPW larvae and adults after 10 days of treatment

Fungi	Concentration (conidia/mL)	Mortality rate(%)
		Larvae	Adult
Commercial formulation	1.0 × 1010	51.11 ± 2.22bc	80.00 ± 5.77ab
	1.0 × 109	42.22 ± 4.41bc	30.00 ± 5.77ef
	1.0 × 108	20.00 ± 3.85d	16.67 ± 8.82fg
Zeb-01	1.0 × 1010	85.00 ± 7.64a	86.00 ± 2.45a
	1.0 × 109	63.33 ± 4.41b	74.00 ± 4.00ab
	1.0 × 108	58.33 ± 1.67b	60.00 ± 5.48bcd
	1.0 × 107	48.33 ± 3.33bc	56.00 ± 4.00cd
	1.0 × 106	35.00 ± 5.00cd	50.00 ± 5.48de
Control		00.0 ± 0.00e	2.00 ± 2.00g
F		39.10 (F8,18)	36.99 (F8,36)
P		< 0.001	< 0.001

Data are mean ± SE. Different letters within the same column indicate significant differences as determined by the Tukey’s test (P < 0.05).

The pathogenicity of the isolated B. bassiana strain Zeb-01 against RPW larvae and adults was compared to a commercial formulation across a range of conidial concentrations (Fig. 5). Against larvae, Zeb-01 demonstrated consistently superior virulence in a concentration- and time-dependent manner. At the highest concentration (1.0 × 10¹⁰ conidia/ mL), Zeb-01 caused significantly higher mortality than the commercial product from day 5 onward. This enhanced efficacy was also clear at lower concentrations; Zeb-01 induced mortality three days earlier and achieved final (day 10) mortality rates of 63.33% and 58.33% (at 1.0 × 10⁹ and 1.0 × 10⁸ conidia/mL), substantially exceeding the commercial product’s rates of 42.22% and 20.00%, respectively. For adults, the two strains exhibited distinct infection kinetics. The commercial product caused rapid initial mortality (30% by day 1 at 1.0 × 10¹⁰ conidia/mL), while Zeb-01 showed a delayed onset (< 2% mortality by day 3).

Fig. 5.

Cumulative mortality of RPW larvae and adults following treatment with the Zeb-01 strain and a commercial (Com) formulation at different conidial concentrations over times. Error bars represent the standard error (± SE) based on three replicates for larvae and five replicates for adults. Different letters indicate significant differences among treatments according to Tukey’s test (P < 0.05)

However, Zeb-01’s virulence increased sharply after day 5. By day 10, it not only matched but significantly surpassed the commercial strain at intermediate concentrations, achieving 74.00% and 60.00% mortality (at 1.0 × 10⁹ and 1.0 × 10⁸ conidia/mL) compared to 30.00% and 16.67%, respectively, highlighting its high ultimate efficacy. Collectively, these findings demonstrate that Zeb-01 possesses superior ultimate virulence against both life stages of RPW, particularly at intermediate and lower conidial concentrations. While the commercial formulation may act more rapidly during early adult infection, Zeb-01 delivers stronger long-term mortality outcomes. Such characteristics suggest that Zeb-01 could serve as a promising biocontrol candidate, especially in IPM programs where sustained efficacy and performance at reduced application rates are critical.

Exposure of RPW larvae to increasing concentrations of strain Zeb-01 resulted in a significant reduction of the LT₅₀ demonstrating a clear dose-dependent virulence response (Table 3). The LT₅₀ values decreased progressively from 12.38 days at 1.0 × 10⁶ conidia/mL to 6.83 days at the highest concentration of 1.0 × 10¹⁰ conidia/mL. This inverse linear relationship between conidial concentration and LT₅₀ was strongly supported by high correlation coefficients (The r-values at all concentrations were approximately 0.99. The closer the absolute value of the correlation coefficient is to 1, the more reliable the derived LT₅₀). Comparative analysis unequivocally demonstrated the superior efficacy of Zeb-01. At 1.0 × 10¹⁰ conidia/mL, the LT₅₀ for Zeb-01 (6.83 days) was significantly shorter (a 63.69% reduction) than that of the commercial strain (11.18 days, r = 0.90). The dose-response relationship was statistically significant at higher concentrations. Non-overlapping 95% confidence intervals at 1.0 × 10¹⁰ and 1.0 × 10⁹ conidia/mL confirmed distinct mortality rates, whereas overlapping intervals at lower concentrations (1.0 × 10⁸ to 1.0 × 10⁶ conidia/mL) indicated non-significant differences in LT₅₀ values. This confirms the pronounced virulence of strain Zeb-01 is concentration-dependent.

Table 3.

Median lethal time (LT₅₀) of RPW larvae after treatment with different concentrations of Zeb-01 strain and commercial B. bassiana, as determined by regression analysis

Fungi	Concentrations (Conidia/mL)	LT50 (days)	Correlation coefficient r	95% Confidence interval
				Lower	Upper
Commercial formulation	1.0 × 1010	11.18	0.90	8.99	13.91
	1.0 × 109	12.96	0.92	9.56	17.57
	1.0 × 108	22.83	0.95	8.76	58.75
Zeb-01	1.0 × 1010	6.83	0.99	6.41	7.27
	1.0 × 109	8.75	0.99	7.85	9.76
	1.0 × 108	8.90	0.99	8.22	9.63
	1.0 × 107	9.96	0.99	9.05	10.96
	1.0 × 106	12.38	0.99	10.26	14.93

The correlation coefficient (r) reflects the strength of the linear relationship between spore concentration and probit-transformed mortality. Values near 1 indicate a strong fit and reliable LT₅₀ estimates. The 95% confidence intervals correspond to the LT₅₀ values calculated for each fungal spore concentration. Narrower 95% confidence intervals denote higher precision. In this experiment, 20 larvae were treated per concentration with three biological replicates.

Adult RPW demonstrated significantly greater tolerance to Zeb-01 infection than larvae, as shown by higher LT₅₀ values across all concentrations (Table 4). Although a dose-response trend was observed (LT₅₀ decreasing from 10.92 to 7.95 days), it was less pronounced than in larvae. The extensive overlap in 95% confidence intervals between concentrations indicates that these differences were largely not statistically significant, revealing a more gradual and less sensitive response in adults. In contrast, the commercial strain exhibited a starkly different virulence profile against adults. It showed rapid efficacy at the highest concentration (LT₅₀ = 2.15 days at 1.0 × 10¹⁰ conidia/mL) but its potency diminished drastically at lower doses (LT₅₀ = 66.99 days and 565.22 days at 1.0 × 10⁹ and 1.0 × 10⁸ conidia/mL). Furthermore, the confidence intervals of these two estimated values were quite wide, probably due to the low mortality rate of the commercial formulation against adults at low concentrations and the relatively long time required for adult mortality to occur after treatment. However, Zeb-01 maintained consistently shorter LT₅₀ values across the same concentration range. Results demonstrate stage-dependent susceptibility in RPW and confirm Zeb-01’s superior consistency and potential compared to the commercial product.

Table 4.

Median lethal time (LT₅₀) of adult RPW after treatment with different concentrations of Zeb-01 strain and commercial B. bassiana, as determined by regression analysis

Fungi	Concentration (Conidia/mL)	LT50 (days)	Correlation coefficient r	95% Confidence interval
				Lower	Upper
Commercial formulation	1.0 × 1010	2.15	0.94	1.29	3.58
	1.0 × 109	66.99	0.96	4.69	956.88
	1.0 × 108	565.22	0.74	0.66	485621.00
Zeb-01	1.0 × 1010	7.95	0.96	7.27	8.69
	1.0 × 109	8.35	0.98	7.72	9.02
	1.0 × 108	10.21	0.96	8.97	11.63
	1.0 × 107	10.82	0.97	9.37	12.48
	1.0 × 106	10.92	0.92	9.39	12.68

The correlation coefficient (r) reflects the strength of the linear relationship between spore concentration and probit-transformed mortality. Values near 1 indicate a strong fit and reliable LT₅₀ estimates. The 95% confidence intervals correspond to the LT₅₀ values calculated for each fungal spore concentration. Narrower 95% confidence intervals denote higher precision. In this experiment,10 adults were treated per concentration with five biological replicates.

The virulence of strain Zeb-01 against RPW larvae was both time- and concentration-dependent (Table 5). Longitudinal analysis showed a progressive decline in LC₅₀ values from 6 to 10 days post-inoculation, with Zeb-01 demonstrating markedly superior efficacy over the commercial strain at every time point. This advantage was most evident on day 6, when the LC₅₀ of Zeb-01 (1.59 × 10¹¹ conidia/mL) was 66,667-fold lower than that of the commercial strain (1.06 × 10¹⁶ conidia/mL). By day 10, Zeb-01 maintained a significantly lower LC₅₀ (1.91 × 10⁷ conidia/mL) compared to the commercial product (5.95 × 10⁹ conidia/mL). Overall, Zeb-01 achieved equivalent mortality at concentrations at least 3000-fold lower than the commercial strain, conclusively demonstrating its superior virulence.

Table 5.

Median lethal concentration (LC₅₀) of RPW larvae after treatment with different concentrations of Zeb-01 strain and commercial B. bassiana, as determined by regression analysis

Fungi	Times (days)	LC50 (conidia/mL)	Correlation coefficient r	95% Confidence interval
				Lower	Upper
Commercial formulation	6	1.06 × 1016	0.87	1.20 × 103	9.31 × 1028
	7	4.94 × 1014	0.87	5.32 × 105	4.58 × 1023
	8	4.93 × 1011	0.98	2.28 × 108	1.07 × 1015
	9	1.92 × 1010	0.98	1.24 × 109	2.99 × 1011
	10	5.95 × 109	0.96	1.17 × 109	3.02 × 1010
Zeb-01	6	1.59 × 1011	0.97	7.30 × 109	3.47 × 1012
	7	1.73 × 1010	0.99	1.54 × 109	1.95 × 1011
	8	2.07 × 109	0.97	3.46 × 108	1.23 × 1010
	9	1.13 × 108	0.97	3.17 × 107	4.05 × 108
	10	1.91 × 107	0.97	5.75 × 106	6.33 × 107

The correlation coefficient (r) reflects the strength of the linear relationship between spore concentration and probit-transformed mortality. Values near 1 indicate a strong fit and reliable LC₅₀ estimates. The 95% confidence intervals correspond to the LC₅₀ values calculated for different treatment time. Narrower 95% confidence intervals denote higher precision. In this experiment, 20 larvae were treated per concentration with three biological replicates.

The temporal virulence profile of strain Zeb-01 against adult RPW exhibited distinct biphasic kinetics, characterized by an initial lag phase followed by a pronounced late-stage virulence enhancement (Table 6).

Table 6.

Median lethal concentration (LC₅₀) of adult RPW after treatment with different concentrations of Zeb-01 strain and commercial B. bassiana, as determined by regression analysis

Fungi	Times (days)	LC50 (conidia/mL)	Correlation coefficient r	95% Confidence interval
				Lower	Upper
Commercial formulation	6	4.57 × 109	0.98	1.77 × 109	1.18 × 1010
	7	3.64 × 109	0.99	1.55 × 109	8.56 × 109
	8	2.81 × 109	0.98	1.23 × 109	6.43 × 109
	9	2.46 × 109	0.98	1.14 × 109	5.30 × 109
	10	2.01 × 109	0.96	9.35 × 108	4.32 × 109
Zeb-01	6	2.11 × 1013	0.99	9.21 × 109	4.83 × 1016
	7	7.49 × 1013	0.90	2.47 × 109	2.27 × 1018
	8	1.41 × 1011	0.87	7.11 × 108	2.79 × 1013
	9	1.87 × 108	0.95	3.18 × 107	1.10 × 109
	10	2.49 × 106	0.91	2.51 × 105	2.47 × 107

The correlation coefficient (r) reflects the strength of the linear relationship between spore concentration and probit-transformed mortality. Values near 1 indicate a strong fit and reliable LC₅₀ estimates. The 95% confidence intervals correspond to the LC₅₀ values calculated for different treatment time. Narrower 95% confidence intervals denote higher precision. In this experiment, adults were treated per concentration with five biological replicates.

Longitudinal assessment revealed a progressive decrease in LC₅₀ values over time, confirming clear time-dependent pathogenicity. During the early infection phase (6–8 days post-inoculation), Zeb-01 showed comparatively higher LC₅₀ values, indicating delayed initial action relative to the commercial formulation. The LC₅₀ values of adults were estimated based on the adult mortality rates observed in the experiment, which were attributed to the low adult mortality rates across all treatments. However, in the late phase (9–10 days post-treatment), strain Zeb-01 displayed significantly enhanced virulence, achieving LC₅₀ values (1.87 × 10⁸ and 2.49 × 10⁶ conidia/mL) that were 10.7-fold and 807-fold lower, respectively, than those of the commercial strain (2.01 × 10⁹ conidia/mL). This biphasic pattern highlights superior persistence of Zeb-01, likely linked to enhanced cuticular penetration and/or immune evasion supporting sustained pathogenicity.

Discussion

Seven microbial isolates (Four fungal and three bacterial) were recovered from naturally infected O. arenosella cadavers and evaluated for pathogenicity against R. ferrugineus under laboratory conditions. Among these, a single fungal isolate, identified as B. bassiana (Zeb-01), exhibited relatively higher pathogenic activity compared to the other isolates obtained in this study. This observed activity may reflect adaptation to local host pathogen interactions, supporting the rationale for isolating strains from ecologically related hosts such as O. arenosella, which co-occurs with RPW in palm agroecosystems (Abrar et al. 2023; Sutanto et al. 2021). B. bassiana infects hosts through cuticular penetration, producing enzymes and toxins that suppress host immunity, a process that may be particularly effective in locally adapted strains (Rajula et al. 2020; Hong et al. 2024). Nevertheless, these findings represent an initial assessment of a single native isolate under controlled laboratory conditions. While the results are encouraging, they should be interpreted within the scope of this exploratory study. Further investigations under semi-field and field conditions are required to confirm the practical efficacy of Zeb-01, as environmental factors such as temperature, humidity, and UV radiation can strongly influence fungal persistence and virulence. Additionally, evaluations of formulation stability, environmental persistence, and non-target safety are necessary before considering this isolate for integration into IPM programs. Future research involving comparative analyses with other B. bassiana isolates from diverse geographic origins or host sources will help contextualize the virulence and ecological adaptability of Zeb-01 within the range of intraspecific variability, clarifying whether its observed activity reflects strain-specific characteristics or typical performance within the species.

Accurate species identification is fundamental for the development of EPF as biocontrol agents (Furuie et al. 2022; Reay et al. 2008). In this study, a fungal strain recovered from cadavers of O. arenosella was initially identified as Beauveria based on morphological characteristics, which showed close resemblance to B. bassiana. However, given the morphological similarities among species within this genus, and the fact that even closely related species can exhibit nearly identical features (Khonsanit et al. 2020), molecular characterization was necessary for definitive identification. The ITS region of the ribosomal RNA gene cluster and a fragment of the translation elongation factor 1-alpha (EF1-α) gene both widely used molecular markers for fungal identification were sequenced. The ITS sequence (587 bp) and EF1-α sequence (508 bp) showed high similarity to previously reported B. bassiana strains (accession numbers OK331343 for ITS; PP466315 and PP466316 for EF1-α), providing molecular confirmation that the isolate belongs to B. bassiana and was designated as strain Zeb-01.

The pathogenic process of B. bassiana is well characterized, beginning with spore germination and formation of appressoria that facilitate penetration of the insect’s cuticle. Once inside, the fungus proliferates, absorbing nutrients from the host and releasing toxic metabolites, including proteases, lipases, and secondary metabolites such as beauverolides, which contribute to host mortality (Hong et al. 2024). B. bassiana is widely used to target pests in agriculture and forestry, including RPW, a destructive pest of palm trees and other tropical crops (Islam et al. 2021; Nurashikin-Khairuddin et al. 2022). Several studies have demonstrated its efficacy against RPW, with reports of up to 100% larval mortality following exposure to high conidial concentrations (Alwaneen et al. 2024; El Husseini 2019; Yang et al. 2023). In the current study, the native isolate B. bassiana Zeb-01 exhibited lower LC₅₀ and LT₅₀ values compared to a commercial B. bassiana formulation under equivalent laboratory conditions. Notably, Zeb-01 maintained a relatively stable LT₅₀ profile across different concentrations, a characteristic that may be relevant for its potential use in pest management strategies. This stability could facilitate rotation or integration with other control methods, though further validation under field conditions is necessary to confirm its practical implications. In adult RPW, Zeb-01 showed delayed virulence onset but achieved higher final mortality than the commercial strain. This pattern may reflect differences in cuticle composition between larvae and adults, as the sclerotized cuticle of adult coleopterans can impede fungal penetration and prolong the infection process. However, this interpretation remains speculative and requires further histological or biochemical investigation. Overall, these laboratory findings document the pathogenic activity of the native isolate Zeb-01 against both larval and adult stages of RPW, supporting its consideration as a candidate for further development as a biocontrol agent.

This study evaluated a single native B. bassiana isolate (Zeb-01) recovered from O. arenosella cadavers under laboratory conditions. As with other EPF, the virulence of Zeb-01 is likely influenced by environmental factors such as temperature, humidity, and UV radiation, which affect conidial persistence and germination (Quesada-Moraga et al. 2024). The isolate also exhibits practical limitations typical of fungal biocontrol agents, including potential susceptibility to inactivation during formulation, sensitivity to high temperatures (≥ 30 °C), and reduced environmental persistence due to humidity fluctuations and soil microbial competition. Although Zeb-01 showed considerable in vitro sporulation and pathogenic activity against RPW larvae and adults, further studies are required to evaluate its performance under field conditions, optimize formulation (e.g., via UV protectants), and assess integration with other control methods (Sutanto et al. 2023). Given that only a single isolate was evaluated, future comparative studies involving other B. bassiana isolates from diverse origins are needed to contextualize the virulence of Zeb-01 within intraspecific variability. Such investigations will help determine whether its observed activity represents strain-specific characteristics or typical performance, and will inform its potential incorporation into IPM programs for palm systems.

Conclusion and perspectives

This exploratory study recovered seven microbial isolates (Four fungal and three bacterial) from naturally infected O. arenosella, among which a single isolate was identified as B. bassiana (Zeb-01). Within the scope of this preliminary laboratory evaluation, Zeb-01 demonstrated measurable entomopathogenic activity against both larval and adult stages of R. ferrugineus, with mortality positively associated with dose and exposure time. Compared with the other isolates obtained in this investigation, Zeb-01 exhibited comparatively higher virulence under the tested conditions. These findings provide preliminary evidence supporting the continued investigation of this native isolate as a potential biological control agent. However, laboratory efficacy does not necessarily translate to field performance. Further research should therefore include field validation, formulation optimization, and assessment within IPM frameworks. Comparative analyses involving additional B. bassiana isolates from diverse origins will be necessary to determine whether the observed activity of Zeb-01 reflects distinct strain-specific characteristics or falls within expected intraspecific variability. Such studies will be important for evaluating its practical suitability in palm pest management systems.

Author contributions

B. Z. Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Resources; Software; Visualization; Roles/Writing - original draft; and Writing - review & editing and C. L. Conceptualization; Funding acquisition; Investigation; Project administration; Resources; Supervision; Visualization and T.M. A. Conceptualization; Data curation; Formal analysis; Investigation; Software; Validation; Visualization; Roles/Writing - original draft; and Writing - review & editing. All authors reviewed the manuscript.

Funding

This research was funded by the National Key Research and Development Program of China (Grant number: 2023YFD2200704).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Competing interests

The authors declare no competing interests.