Construction Of Engineered Beauveria Bassiana Strain Expressing The Vegetative Insecticidal Protein Vip3Aal Of Bacillus Thuringiensis And Its Oral And Cuticular Infectivity To Spodoptera Litura Larvae

As a classic entomopathogenic fungus, Beauveria bassiana usually infects insects through cuticle and is characteristic of slow lethal effect due to a latent period of several days. For this reason, fungal biocontrol agents are usually less effective against foliage feeder pests with chewing mouthparts. Virulence improvement of fungal candidate strains is often achieved by the integration of endogenous cuticle-degrading enzymes or exogenous insecticidal proteins into fungal genome but rarely by the introduction of intestine-virulent factors. Bacillus thuringiensis (Bt) enables to secrete vegetative insecticidal proteins, such as Vip3A, which are highly toxic to lepidopteran pests after ingestion. However, low-level secretion and non-crystal instability of such toxins make them difficult to be formulated like Bt endotoxins for field application except a mere approach to expressing them in transgenic plants for insect resistance. Making use of easily produced and formulated fungal conidia as Vip3A vectors for dual infection through cuticle and ingestion would increase a chance of host infection. This is a new challenge for microbial control of foliage feeder pests.The present study began from selecting a B. bassiana strain with high virulence against Spodoptera litura larvae, followed with transforming vip3Aal gene (one of Vip3A-coding genes) into the selected strain. Biologically robust bioassays were conducted with the wild-type strain and the engineered strain expressing Vip3Aal to evaluate their infectivity to S. litura larvae through cuticle, ingestion or both. The results are summarized below.Screening of B. bassiana isolates against S. litura larvae. To select a B. bassiana isolate as a recipient of Vip3Aal,12 isolates with desired features of growth, conidiation and toleration to environmental stresses were separately sprayed onto the second-instar larvae of S. litura with 2 ml of conidial suspension (standardized to 1×108 conidia/mm2) using an Automatic Potter Spray Tower. A blank control was sprayed only with 0.02% Tween-80. After spray, all larvae were reared at the regime of 25℃and 12:12 (L:D) and monitored daily for 7-day mortality records. As a result, the corrected mortalities of the treated larvae (87-97 larvae per treatment of three replicates) ranged from 33.9 to 82.4% for the 12 isolates under the mean spray of 2473 conidia/mm2. Median lethal time (LT50) of 10 isolates causing>50% mortalities ranged from 3.9 to 5.4 days. Of those, the isolate Bb2860 was most infective to the larvae and thus selected as a recipient strain of the vip3Aal gene.Detection and insecticidal activity of Vip3Aal expressed in transgenic B. bassiana. The vip3Aal gene was transformed into the wild-type strain Bb2860 (BbW) by means of blastospore-based transformation system with the phosphinothricin (PPT) resistance gene bar as selective marker. A genetically stable transformant with growth and conidiation features as usual was obtained by screening through PCR, RT-PCR and Western blotting. This transformant, named as BbV28, showed excellent constitutive expression of Vip3Aal in both mycelia and conidia in Western blot. In immunogold localization, the expressed Vip3Aal were abundant and distributed evenly in the conidial cytoplasm of BbV28. The fourth-instar S. litura larvae were fed with cabbage leaf discs immerged in the Bbv28 suspension of 5x108 conidia/ml or the lysate of 60μg/ml Vip3Aa expressed in E. coli cells and showed similar symptom of shrink and palsy after for 18 h feeding. In contrast, such a symptom did not appear in the larvae fed with leaf discs treated with the same suspension of BbW conidia or 0.02% Tween-80 at the same time. The midgut juice samples of the larvae feeding on the BbV28-treated leaf discs for 18 and 36 h were positively blotted with the polyclonal antibody of Vip3Aal in Western blot, yielding distinct bands similar to those from the midgut samples of the larvae feeding for the same periods on the leaf discs treated with Vip3Aa expressed in E. coli. The main band detected was-62 kDa, which is well in agreement with the molecular size of the active fragment of midgut-digested Vip3A in previous reports. However, no band appeared in the profiles of the midgut juice samples from the larvae fed with BbW-treated leaf discs. The results indicate that the Vip3Aal expressed in BbV28 conidia can be released into larval midgut to act as an intestine virulence factor after ingestion.Comparative virulence of BbV28 and BbW against S. litura larvae through oral and/or cuticular infection. Both BbV28 and BbW were bioassayed for their virulence to the second-instar larvae of S. litura through the infection of cuticle (AssayⅠ), ingestion (AssayⅡ) or both (AssayⅢ). Each assay included the three 1-ml sprays of 4×106,2×107 and 1×108 conidia/ml plus 0.02% Tween-80 as blank control under the spray tower. Thus, the three fungal sprays in each assay generated low, median and high concentrations of 53-66,337-390 and 1242-1397 conidia/mm2, respectively. All assays were repeated three times with 30-40 larvae per treatment at each time. After spray, the larvae were fed with daily changed fresh cabbage leaves free of conidia (Assay I) or sprayed with a given conidial suspension (AssaysⅡandⅢ) at the same regime. Mortality in each treatment was daily examined for 7 days and cadavers found at each time were incubated at saturated humidity for fungal outgrowths. All observed mortalities were subjected to two-factor (strain and conidial concentration) analysis of variance (ANOVA) and fitted to the time-concentration-mortality (TCM) model with the DPS software.As a result, larval mortalities generally increased with conidial concentration and post-treatment time length for both fungal strains in AssaysⅠ-Ⅲ. The cumulative mortalities caused by both strains differed significantly from one bioassay to another or among the three conidial concentrations. BbV28-treated larvae tended to die faster than those treated with BbW at a given concentration in AssaysⅡandⅢbut this difference was not present in Assay I. The control mortalities (3.8-8.1%) were similar in the three assays. Moreover, all larvae killed by BbV28 and BbW in Assay I showed typical symptom of mycosis, being well mycotized after 3-day incubation at saturated humidity. In AssayⅡorⅢ, the larvae killed earlier by BbV28 were often shrunk and smaller, being poorly mycotized under saturated humidity, but those died of BbW exhibited the same symptom as seen in AssayⅠ.All the TCM trends of both strains in AssaysⅠ-Ⅲwere well fitted to the TCM model. The fitted TCM relationships were used to calculate the trends of LC50s over the post-spray time lengths and of LT50s over the conidial concentrations for the two strains. The LC50s and associated 95% confidence limits of BbW and BbV28 against the larvae directly exposed to the fungal sprays in Assay I were estimated as 2.6 (0.8-8.5)×104 and 1.5 (0.6-4.1)×104 conidia/mm2 on day 3 and declined to 750 (571-985) and 698 (539-900) conidia/mm2 on day 7, respectively. These estimates were not significantly different between the two strains. However, the LC50 estimates differed significantly between BbV28 and BbW in AssaysⅡandⅢ. Compared to AssayⅠ, the ingestion of BbV28 conidia by S. litura larvae in AssaysⅡandⅢled to the respective LC50 reductions of 26.2 and 17.2 fold on day 3 and of 1.1 and 1.3 fold on day 7. On the other hand, the LT50 trends of both strains decreased similarly over the conidial concentrations in AssayⅠ. For example, a concentration of 1000 conidia/mm2 resulted in the LT50s of 5.56 and 5.61 days for BbV28 and BbW, respectively. In contrast, the LT50s of BbV28 were shortened by 26.9-34.8% at 800-1500 conidia/mm2 in AssayⅡand by 23.5-35.3% at 500-1500 conidia/mm2 in AssayⅢ. Conclusively, the virulence of the transgenic strain BbV28 to S. litura larvae was greatly enhanced because it enabled to infect the larvae not only through cuticle as usual but also through ingestion.