Molecular characterization and virulence of fungal isolates against the bean flower thrips, Megalurothrips usitatus Bagnall (Thysanoptera: Thripidae)

The bean flower thrips, Megalurothrips usitatus Bagnall (Thysanoptera: Thripidae) is a major pest of leguminous crops grown in Southern China. This study reports the isolation, identification (through molecular characterization) and pathogenic potential of 6 fungal isolates belonging to different fungal species. The fungi were isolated from soil samples collected from different areas of Southern China and were identified as Beauveria bassiana (3 isolates; SB010, SB009, SP016), Cordyceps fumosorosea (one isolate; SP535), Akanthomyces attenuatus (one isolate; SCAUDCL-53), and Aspergillus nomius (one isolate; SCAUN-1). Conidial suspension (1 × 108 conidia ml−1), and the crude extract (0.4 mg ml−1) of all the 6 fungal isolates were tested for their pathogenicity against M. usitatus adults. The results showed that A. attenuatus (SCAUDCL-53) and B. bassiana (SB010) were the most effective fungal isolates against M. usitatus out of all the isolates used in this study. At 5 days post-inoculation, conidial suspension of A. attenuatus (SCAUDCL-53) and B. bassiana (isolate SB010) caused 100 and 90% mortality rates, respectively. The median lethal time (LT50) values of fungal isolates SCAUDCL-53, SB010, SB009, SP016, SP535, and SCAUN-1against M. usitatus adults were 1.36, 3.79, 6.51, 8.49, 17.36, and 5.01 days, respectively. The application of crude fungal extracts of SCAUDCL-53, SB010, SB009, SP016, SP535, and SCAUN-1 against the pest resulted in 85, 93.3, 56.7, 33.3, 41.7, and 53.75% mortality rates, respectively after 5 days of application. Respective LT50 values of the crude fungal extracts against M. usitatus were 3.37, 2.85, 4.87, 7.13, 6.43, and 4.64 days. The fungal isolates used in this study showed considerable bioactivity against the M. usitatus and can be used as potential natural pest control agent for the ecofriendly management of M. usitatus.


Background
The bean flower thrips, Megalurothrips usitatus Bagnall (Thysanoptera: Thripidae), is a major pest of cowpea causing considerable economic losses to this crop by damaging cow pea leaves, flowers and pods (Mound and Walker, 1987). M. usitatus damage can cause leaf wrinkling, growth point atrophy, pod scab, and other effects on cowpea plant reducing the market value of crop (Tang et al. 2015). The flowering stage is known to be the most vulnerable to M. usitatus attack and hundreds of M. sitatus (nymphs or adults) can be seen per flower during peak pest ultimately leading to reduced yield through premature flower loss (Tang et al. 2015). Currently, chemical insecticides are being used as the principal control agent against M. usitatus. They are not effective enough due to the cryptic habits of M. usitatus (such as staying within flowers and their short life cycles) (Tang et al. 2015). The injudicious use of chemical insecticides has resulted in development of insecticide resistance in M. usitatus (Immaraju et al. 1992). Therefore, the development of ecologically sound and sustainable control measures for M. usitatus management is required on urgent basis (Elimem et al. 2018).
Entomopathogenic fungi (EPF) have received considerable attention as biological control agents of thrips and several studies have demonstrated the effectiveness of them (such as Beauveria bassiana, Metarhizium anisopliae, Paecilomyces fumosoroseus, A. attenuatus, Aschersonia aleyrodis) against different thrips species (Skinner et al. 2012 andWright et al., 2016). Ugine et al. (2006) reported that application of B. bassiana significantly reduced Frankliniella occidentalis populations by 81%, especially at the flowering stage of crops. The Verticillium lecanii (V3450, Vp28) is a well-known species of EPF against thrips and some isolates of this species are being developed as commercial biopesticides (Ainsworth et al., 2008).
The potential use of B. bassiana, C. fumosorosea, and A. attenuates have been demonstrated on many insect pests (Ali et al. 2018 andZimmermann 2018). And the insecticidal attribute of B. bassiana and V. lecanii developed its use in biopesticide industry (Singh et al. 2015). Isaria fumosorosea is mainly used to control agricultural pests such as Bemisia tabaci, aphids and thrips. A. nomiu is a facultative pathogen of many plants and animals (St Leger et al., 2000).
The present study was designed to isolate, identify, and describe different strains. The fungal conidia and crude extract were also tested for their pathogenicity against M. usitatus.

Sample collection and fungal isolation
The soil samples were collected from farmers' fields from 6 localities (Yunan, Hunan, Guangxi, Guangdong, Fujian and Hainan) of Southern China. Fungal isolation was carried out following the method described by Gottel et al. (2008) and Imoulan et al. (2011). Briefly, 3 g soil sample was added to 30 ml of ddH 2 O containing 0.05% Tween 80. The mixture was vortexed for 15 min and 1 ml suspension was layered onto potato dextrose agar (PDA) plates. The plates were incubated at 26 ± 6°C, 70 ± 5% RH and 16:8 h L:D photoperiod for 7 days. The growth of the fungi was observed after 7 days and were re-inoculated by inoculation of individual fungus on new PDA plates. In this way, several rounds of inoculation were performed until a purified strain based on the phenotypic properties and fungal morphology was obtained (Saito andBrownbrideg, 2016 andDu et al. 2019) DNA extraction, PCR amplification, and construction of phylogenetic trees Genomic DNA of fungal isolates, purified after the isolation process, was extracted by using a fungal genomic DNA extraction kit supplied by Ezup, Sangon Biotech, Shanghai, China Internal transcribed spacer region (ITS) of fungal DNA was amplified in order to identify the species of each strain (Imoulan et al. 2016). The reaction mixture for PCR amplification consisted of 25 μl 2 × TapPCR Master Mix, 1 μl of each primer (10 μM), 1 μl genomic DNA, and 22 μl ddH 2 O. The ITS regions were amplified using the primers ( Table 1). The thermal profile of the PCR condition for primer included an initial denaturation at 94°C for 5 min, followed by 35 cycles with a new denaturation at 94°C for 30 s, annealing at 53°C for 30 s, extension at 72°C for 1 min, and with a final extension of 10 min at 72°C. The PCR products were assayed for fragment size by agarose gel electrophoresis. Then, the PCR products were sent to the Shanghai Majorbio Bio-pharm Technology (Shanghai, China) for sequence analysis. The obtained ITS gene sequences of all strains were sequence-spliced with Genious version 7.1.4 (Goloboff and Catalano, 2016), then aligned in National center for biotechnology information (NCBI), the aligned sequences were downloaded, and then all sequences were constructed with MEGA Version 7.0 to construct a maximum likelihood (ML) tree (Kumar et al. 2016).

Insect colony
Adults of M. usitatus were collected from a cowpea field at South China Agricultural University, Guangzhou, China during 2017. The collected insects were subsequently reared under laboratory conditions by the bean pod method (Espinosa et al. 2005). The insect colonies were kept at 26 ± 6°C, 70 ± 5% RH and 16:8 h L:D photoperiod in a climate control chamber. Newly emerged adult females were used for fungal bioassay studies.

Fungus culture, conidia suspension, and crude extract preparation
The tested strains (SCAUDCL-53, SP016, SP535, SB009, SB010, and SCAUN-1) were cultured on PDA plates for 7 days under laboratory conditions. Conidia were Fifty milliliters of freshly prepared sabouraud dextrose broth medium (SDB) was added to 250-ml Erlenmeyer flasks followed by sterilization at 121°C for 30 min. Five milliliters of freshly prepared fungal suspension (1 × 10 7 conidia ml −1 ) of different fungal isolates was added to culture medium. The flasks were incubated in a rotary shaker at 120 rpm, 28°C for 5 days. After 5 days, fungal cultures were filtered through Whatman filter paper (Qian Yihua Glass Instruments Co. Ltd) to obtain the supernatant for subsequent studies. The protein concentrations of crude extracts were quantified by using bouvine albumin serum as substrate (Bradford 1976). The protein concentration of all the crude extracts was adjusted to 0.4 mg mL −1 by following Quesada et al. (2006).

Bioassay method
Laboratory bioassays were conducted to evaluate the toxicity of different fungal isolates identified during this study against the M. usitatus adults by using centrifuge tube residual bioassay (Du et al. 2019). The centrifuge tubes (9 ml) and bean pods (1 cm) were immersed in conidial suspension (1 × 10 8 conidia ml −1 ) and crude extract for 2 h, followed by drying under sterilized conditions. The centrifuge tubes and bean pods treated with ddH 2 O having 0.05% Tween-80 served as control. After drying, adult females of M. usitatus (100 individuals) were transferred to the tubes having bean pods treated with same fungal treatment. Each centrifuge tube was sealed with a cotton plug to prevent thrips from escaping. The whole experiment was repeated thrice with fresh batches of insects at 26 ± 1°C, 75% RH, and 16:8 L:D photoperiod. The insects were observed on daily basis to record the mortality data following Du et al. (2019).

Microscopic examination
Newly emerged M. usitatus adults were inoculated by the conidial suspension (1 × 10 8 conidia mL −1 ) and the crude protein extract (0.4 mg ml −1 ) of all the 6 fungal isolates and were incubated at 25°C and 75% RH in the dark. The infected M. usitatus (5 individuals) were sampled at 1, 2, 3, 4, and 5 days after inoculation. Gross changes in the appearance of the infected M. usitatus were directly monitored at different times after inoculation under a D850 camera (Nikon Co. Ltd. Japan).

Statistical analysis
Mortality data were percent transformed and were subjected to probit analysis. All the analysis were performed through SAS software v9.1 (SAS et al., 2000).

Identification of the fungi
The present study reported 6 fungal isolates belonging to 4 fungal species from isolated soil samples collected from different localities of southern China. The result showed that the strains SB009, SB010, and SP016 were identified as B. bassiana (NR_111594). Strain SP535 was identified as C. fumosorosea (AF461747). Strain SCADDCL-53 was consistent with the A. attenuatus; MH558279 and isolate SCAUN-1 was identified as A. nomius (NR_121218) (Fig. 1).
The identification of entomopathogenic fungi, based on morphological features, can lead to ambiguous results (Ziółkowska et al. 2015). Nucleic acid sequence analysis is commonly used method for the identification and classification of fungi (Funk et al. 1983). It compares the correlation between homologous molecules by determining the composition of nucleotide sequences in the primary structure of nucleic acid (Diaz et al. 2012). Sequencing of the ITS region of rRNA is currently regarded as the standard method for phylogenetic analyses and identification of fungal species (Schmidt and Moreth, 2002;Cafarchia et al. 2013 andKawasaki 2011) (Table 2). In this study, the comparison of ITS sequences of the isolated fungi showed that they were closely related to B. bassiana (isolates SB009, SB010, SP016), C. fumosorosea (isolate SP535), A. attenuatus (isolate SCAUDCL-1), and A. nomius (isolate SCAUN-1).    Ravensberg et al. (1990) reported that sprayings V. lecanii at weekly intervals reduced the F. occidentalis incidence by 90%. Wang (2012) screened out different isolates of Lecanicillium lecanii for the management of Gynaikthrips ficorum and showed more than (85.56%) mortality of G. ficorum by L. lecanii isolateV16063. The LT 50 values (1.36 days) of A. attenuatus isolate SCAUDCL-53 against M. usitaus at a concentration of 1 × 10 8 conidia mL −1 observed during this study are a little lower than those observed by Du et al. (2019) who observed LT 50 of 3.5 days for A. attenuatus isolate SCAUDCL-38. B. bassiana is known as an effective biological control agent of several insect species (Rezende et al. 2009). We observed high pathogenicity of B. bassiana (SB010) towards M. usitaus under laboratory conditions, suggesting that it has high capacity in M. usitaus control. In the current study, conidial suspension (1 × 10 8 conidia mL −1 ) of B. bassiana isolate SB010 exhibited 90% mortality of M. usitaus after 5 days of fungal application whereas the median lethal time (LT 50 ) of conidial suspension was 3.79 days. However, the isolates SB009 and SP016 of B. bassiana just inhibited 36.25% and 36.21% of M. usitaus. Jacobson et al. (2001) reported that B. bassiana showed remarkable pathogenicity on F. occidentalis and reported 87% mortality. The differences in the values of mortality compared to those obtained in the present study could be explained by the different B. bassiana species and experimental insects hosts (Jacobson et al., 2001). At present, I. fumosorosea is mainly used to control agricultural pests such as whiteflies, aphids, and thrips (Faria and Wraight, 2007;Kabaluk and Gazdik, 2005). In this study, M. usitaus were less susceptible to the strain SP535 isolated of C. fumosorosea, showing 25% mortality.
Although fungal conidia are commonly used to control insect pests, this application method has limitation as  fungal conidia can be susceptible to environmental conditions (Skinner et al. 2012;Rangel et al. 2008

Microscopic examination
The 6 fungal isolates induced similar infection symptoms in M. usitatus adults. In the prophase of postinfection, M. usitaus' head generated white hyphae, along with the post-infection of time, the hyphae developed continuously around the insects' body white hyphae covered the whole insect body and the activities of insects were abnormal (Fig. 4). However, the insects were treated by crude protein extracts generated hyphae earlier than conidial suspension. Furthermore, the density of hyphae was also higher, following the crude protein extracts application when compared with the application of conidial suspension (Fig. 5).

Conclusion
Successful isolation and molecular characterization of 6 different strains belonging to 4 fungal species were reported. The pathogenic potential of these isolates was also observed through bioassays of conidial suspension and crude protein extracts against M. usitatus adults. B. bassiana isolate SB010 and A. attenuatus isolate SCAUDCL-53 were the most effective against M. usitatus. These findings will provide baseline information about the screening of effective biological control agents against M. usitaus. Further field trials as well as studies on identification of effective fungal toxins are still required to design efficient and environmentally sustainable biopesticides for M. usitaus management.