Cry75Aa (Mpp75Aa) Insecticidal Proteins for Controlling the Western Corn Rootworm, Diabrotica virgifera virgifera LeConte (Coleoptera: Chrysomelidae), Isolated from the Insect-Pathogenic Bacterium Brevibacillus laterosporus

Isolation and growth of B. laterosporus strains.B. laterosporus strains EG5553, EG5551, and EG5552 were sourced from grain dust samples obtained from farmers in the eastern and central regions of the United States, and the strains described in this study were isolated in 1986. Grain dust samples were treated as described in Donovan et al. (33, 34) and spread on NYSM agar plates to isolate Bacillus-type colonies. Phase-contrast microscopy of sporulating colonies identified B. laterosporus-type sporangia production with the typical canoe-shaped parasporal body (CSPB) firmly attached to one side of the spore. Isolated B. laterosporus strains were further analyzed by a modified Eckhardt agarose gel electrophoresis procedure (35) to determine the array of native plasmids found in each strain. Strains with unique plasmid arrays and crystal morphologies were selected for further study. The B.O.D. strain was isolated from capsule powder of the commercially available Latero-Flora human dietary supplement. The powder was resuspended in 1 ml of sterile water, heat shocked in an 80°C water bath for 20 min, plated on NYSM agar, and incubated at 28°C for 3 days. Colonies were picked and examined under phase-contrast microscopy to identify B. laterosporus-type sporangia with the typical CSPB firmly attached to one side of the spore (19, 28).

DNA isolation.Cultures of EG5551, EG5552, EG5553, and B.O.D. for DNA extraction were initiated from −80°C glycerol stocks. Each culture was started by adding 100 µl to 900 µl of sterile terrific broth (TB) to a deep-well 96-well plate, covered with an AirPore sheet, and placed in a Multitron shaker at 28°C with shaking at 900 rpm overnight. The next day, 2 wells containing fresh 900 µl TB were inoculated with 100 µl of the overnight culture, covered with an AirPore membrane, and placed back on the Multitron at 28°C and 900 rpm for ∼3 to 4 h. Plates were centrifuged at 2,700 × g for 10 min in an S5700 swinging bucket rotor (Beckman Coulter), the supernatant was discarded, and pellets were washed with 900 µl/well of sterile 1 M NaCl and centrifuged again at 2,700 × g for 10 min and placed at −20°C until DNA isolation.

Qiagen DNeasy blood and tissue DNA kits were used for DNA isolation with several modifications to the protocol described below (Qiagen, Germantown, MD). Bacterial cell pellets were thawed at room temperature for 30 min, and 400 µl of a lysis buffer cocktail consisting of 10 mM Tris-HCl, pH 8.0, 0.5 mM EDTA, pH 8.0, 0.005% Triton X-100, 0.5% Tween 20, 2 µl RNase A (20 mg/ml), and 2 µl Riboshredder was added, fully resuspending bacterial cell pellets. Using a heated shaker platform, samples were incubated at room temperature for 10 min (no shaking), 37°C for 10 min (with slow shaking), and 56°C for 30 min (with slow shaking). Then 200 µl AL buffer (no ethanol added) and 25 µl of proteinase K µg/ml were added to each well. Samples were incubated at 65°C for 10 min with slow shaking, and finally, 350 µl of ethanol was added. From each sample, 900 µl of lysate was added to the kit purification plate. The plate was sealed and centrifuged at 4,300 × g for 30 min. Then, 500 µl/well of AW1 buffer was added to the plate, sealed, and centrifuged for 30 min at 4,300 × g. We then added 500 µl/well of AW2 buffer, and the plate was centrifuged at 43,00 × g for 15 min. The column plate was placed over the elution plate, 160 µl/well of AE elution buffer was added and let sit for 1 min, and then the plate was sealed and centrifuged at 2,700 × g for 10 min. DNA samples were quantified using a Trinean 96-well DNA microfluidics plate reader. To each lane, 2 µl of DNA was added and analyzed for DNA concentration and purity.

Genome sequencing.Total extracted genomic DNA from, EG5553, EG5551, EG5552, and B.O.D. was subjected to massively parallel sequencing on the Illumina HighSeq platform using the 2 × 100-bp paired-end format. DNA sequencing libraries were prepared using the Nextera library prep method (Illumina Inc., San Diego, CA). Reads were assembled using the CLCBio Genomic Workbench or the SPAdes algorithm. Assembled contigs were analyzed using a custom bioinformatic pipeline developed by Bayer Crop Science for gene calling and protein annotation. The bioinformatic analysis included BLAST against public protein databases (Swiss-Prot and Uniref100), Pfam analysis, SignalP, which predicts the presence of membrane-transiting signal peptides and the location of their cleavage (36, 37), and a custom insecticidal protein database curated and maintained at Bayer Crop Science.

Phylogeny construction by 16S rRNA analysis.A phylogeny of select endospore-forming bacilli was constructed using the ribosomal 16S RNA sequence (16S rRNA) (38). The 16S rRNA gene sequences were obtained from NCBI or generated from genome sequences of isolates described in this study with barrnap version 0.8 (https://github.com/tseemann/barrnap). Gene sequences were aligned using cmalign (INFERNAL 1.1.3 suite) (39) using the Rfam model for the small subunit rRNA to guide alignment (RF00177; http://rfam.xfam.org/family/RF00177). The resulting alignment was used to infer a maximum-likelihood phylogeny with FastTree version 2.1.10 SSE3, using the default options (40), and the resulting tree was visualized with ggtree (41).

Mpp75Aa identification, cloning, and expression.The full-length genes encoding TIC3670, TIC3669, and TIC3668 were putatively identified in the draft genome assemblies based on Pfam analysis and BLAST hits to known insecticidal proteins. PCR primers were designed based on the first 30 nucleotides of the 5′ end and the last 30 nucleotides of the 3′ ends of the coding region to generate PCR products. An additional 16 nucleotides were added to both 5′ and 3′ ends to enable hot fusion cloning (42). PCR products were generated using standard PCR conditions with KOD hot start DNA polymerase (Novagen, San Diego, CA). The same DNAs used for the whole-genome sequencing were used to generate PCR products for hot fusion. PCR products were cloned into a B. thuringiensis expression/E. coli shuttle vector (designed and created at Bayer Crop Science). The mature forms of the three proteins (mMpp75Aa; membrane-transiting signal peptide removed) were subcloned into E. coli strain DH10B. Plasmid DNA from E. coli clones containing TIC3668, TIC3669, and TIC3670 (with/without signal peptides) was isolated and sequence confirmed by Sanger sequencing (ABI 3700). The three confirmed protein sequences were submitted to the Bacillus thuringiensis Nomenclature Committee and assigned Cry designations (http://www.lifesci.sussex.ac.uk/home/Neil_Crickmore/Bt/). Recently, the BPPRC reviewed the three proteins and assigned new designations (https://www.bpprc.org/) (2, 27).

B. thuringiensis expression plasmids, bulked up in E. coli, were transformed by electroporation into an atoxigenic B. thuringiensis expression strain for protein production (43). B. thuringiensis expression strains were selected on Luria agar (LA) plates with 5 µg/µl chloramphenicol and colonies picked for protein expression in C2 liquid culture with 5 µg/µl chloramphenicol. Cultures were incubated at 28°C for 3 to 4 days in 1× C2 medium [3.11 g KH₂PO₄, 4.66 g K₂HPO₄, 2.0 g peptone, 2.0 g yeast extract, 2.0 g (NH₄)₂SO₄, 5.0 g casamino acid, 10.0 g glucose dissolved in 900 ml ddH₂O, and 100 ml 10× C2 salts composed of 3.0 g MgSO₄·7H₂O, 1.0 g CaCl₂·2H₂O, 0.58 g MnCl₂·4H₂O, 0.050 g ZnCl₂·7H₂O, 0.050 g CuSO₄·5H₂O, and 0.005 g FeSO₄·7H₂O in 1 liter] and assessed for crystal formation and sporulation. Protein crystals and spores (CSP) were harvested by centrifugation at 8,000 × g; the supernatant was discarded and washed in 10 mM Tris-HCl, pH 8, 0.005% Triton X-100, and 0.1 mM EDTA (pH 8) (TX); and the cell pellet was resuspended in TX at 1/10 the volume of the original culture and stored at 4°C. The insoluble protein in the washed CSPs was solubilized by resuspending the washed cell pellet in 25 mM carbonate buffer (pH 10.0) and 5 mM dithiothreitol (DTT) in 1/10 the original volume followed by centrifugation at 8,000 × g to pellet any remaining insoluble debris. The supernatant was collected and stored at 4°C. Cry75Aa protein quantification for CSPs or solubilized CSP was determined by spot densitometry analysis of SDS-PAGE on a Bio-Rad Gel Doc EZ system.

The genes encoding the TIC proteins were cloned into a pET containing an in-frame C-terminal His tag and E. coli expression vector and transformed into E. coli Bl21 cells for protein production. Cells were grown in auto-induction media at 37°C, and the C-terminal His-tag protein was purified using batch nickel chromatography and buffer exchanged into 25 mM carbonate buffer (pH 10) and 5 mM DTT. Proteins estimated to be greater than 90% pure (by SDS-PAGE) were quantified by Bradford assay and stored at −80°C.

WCR diet overlay bioassay.For each insect bioassay, WCR eggs were stored at 10°C, 60% rH, and 0:24-h light:dark cycle. Eggs were incubated for 13 d at 25°C, 70% rH, and 0:24-h light:dark cycle. On day 13, eggs were washed from soil and surface disinfested (44) Eggs were incubated overnight for hatch. Two hundred microliters WCR artificial diet (44) was dispensed per well in a 96-well plate, cooled in a laminar flow hood, bagged, and stored at 4°C. On the day of egg hatch (day 14), proteins were removed from the −80°C and thawed by placing on wet ice while sample dilutions were prepared. Prior to diet overlay, a solution containing 4 antibiotics was mixed with the samples to achieve a final concentration of 2.6 µg/ml ciprofloxacin, 4.0 µg/ml colistin, 4.0 µg/ml tobramycin, and 5.0 µg/ml amphotericin B. Samples were arrayed in a 96-deep-well block before overlaying on insect diet. Proteins were mixed thoroughly, and 20 µl of sample was overlaid onto diet per well. Each sample per treatment consisted of 3 replicate columns with 8 wells/column (n = 24). Plates were dried in the laminar flow hood before adding one WCR neonate per well with a fine paintbrush. Plates were sealed with VWR silicone adhesive film and ventilated using an “000” insect pin. Bioassay plates were incubated at a 27°C and 70% rH in complete darkness for 6 days and were scored for mortality and instar/stunting as described below. Data were analyzed using JMP 12 statistical software (SAS Institute).

Colorado potato beetle diet bioassay overlay.Colorado potato beetle (CPB) (Leptinotarsa decemlineata) egg masses were held at 25°C until needed and then washed in a 1% bleach and 0.1% liquid soap solution. Eggs were placed on Kimwipes in hatch tubs and incubated at 27°C for 4 days or until eggs hatched. Artificial diet, CPB diet product, catalog no. F9380B (Frontier Scientific Services), was prepared according to the manufacturer’s instructions. Samples were arrayed in a 96-deep-well block before overlaying on insect diet. Proteins were mixed thoroughly, and 20 µl of sample was overlaid onto diet per well. Each sample per treatment consisted of 3 replicate columns with 8 wells/column (n = 24). Neonates were manually infested with one insect per well in a 96-well diet bioassay plate previously overlaid with protein test samples or buffer/water control samples and air-dried. Plates were sealed with preperforated heat seals and incubated at 27°C incubator for 5 days. Column-wise mortality and stunting were evaluated at the end of the assay as described below.

Southern corn rootworm diet bioassay overlay.Southern corn rootworm (SCR) (Diabrotica undecimpunctata howardi) eggs were stored at 10°C. Eggs were separated from soil using clean 30- and 60-mesh sieves. Soil was washed through the sieves with a high-pressure water sprayer, and the clean eggs were collected in a 250-ml beaker. Eggs were sanitized with a 1% bleach solution for 1 min and allowed to settle. Floating debris and dead eggs were decanted followed by a sterile water rinse and decanting and repeated three times. Eggs were washed in 10% formalin and a 0.1% liquid soap solution for 3 min and allowed to settle. Floating debris and dead eggs were decanted followed by a sterile water rinse and decanted; this was repeated three times. SCR eggs were suspended in 0.19% Serva agar solution with a ratio of 1 egg to 20 μl agar and dispensed through a multiple-channel pipettor onto blotting paper. The blotting paper was placed on top of a blank 96-well assay plate and incubated at 25°C with 60 to 70% humidity until hatching. Samples were arrayed in a 96-deep-well block before overlaying on insect diet. Proteins were mixed thoroughly, and 20 µl of sample was overlaid onto diet per well. Each sample per treatment consisted of 3 replicate columns with 8 wells/column (n = 24). Neonates were knocked down into a 96-well diet bioassay plate previously overlaid and air-dried with the protein test samples or buffer/water control samples. The infested bioassay plates were incubated at 25°C for 5 days. Assays were scored for mortality and stunting in a column-wise manner as described below.

Diet bioassay scoring method.Scoring for all diet assays was completed in a column-wise manner. Sample columns were visually compared to the control H₂O column individually. Stunting scores were determined based on the visual difference in size between the sample column and the H₂O column. The stunting score was rated on a 0 to 3 scale. A stunting score of 0, 1, 2, or 3 is defined as less than a <25%, 25 to 50%, 50 to 75%, or >75% difference in size between the sample column and the H₂O column, respectively. Mortality per well was determined if all larvae within a well were dead. Percent mortality was the number of wells with dead larvae from the total 24 wells. The bioassay was discarded if less than 70% of wells/concentration were infested or with more than 15% contamination. Neonates from a single column of wells from the H₂O control were collected and weighed at assay completion to confirm control insects met a minimum weight.

Transgenic plant construction.In planta expression was carried out using both the full-length Mpp75Aa and the mature form of Mpp75Aa (mMpp75Aa; membrane-transiting signal peptide removed) proteins. Codons of the three Mpp75Aas were redesigned for expression in monocots with a GC content of 59%. Mpp75Aas were expressed in a proof-of-concept expression cassette with a high constitutive expression pattern. Cassettes contained a CaMV35S promoter with a duplicated enhancer (45), a 5′ leader sequence from the wheat major chlorophyll a/b-binding gene (46), an intron from the rice actin 1 gene inserted between the leader and coding sequence (47), and a 3′ untranscribed region (UTR) from the wheat 17-kDa heat shock protein gene (48). In the second set of constructs, a plastid-targeting signal peptide from Setaria italica granule-bound starch synthase (49) was fused in-frame with the coding sequences of the mMpp75Aa homologs. All cassettes were cloned into binary vectors containing a CP4_EPSPS selection cassette providing glyphosate resistance. These six cassettes were transformed into the maize LH244 genotype. Single-copy R0 events were selected based on molecular assays and grown to produce hybrid F1 seed by transferring pollen from transgenic R0 events to a 93IDI3 donor ear. F1 seeds were germinated and were sprayed with glyphosate after a week to eliminate transgene-negative segregants.

Growth chamber root protection assay.Seeds from different maize lines were individually germinated or transplanted into 8-in. pots containing Berger BM6 soil in growth chamber at 25°C day/21°C night, 16:8-h light:dark cycle, 50% humidity, and 650 luminescence after surviving a glyphosate application. Protein expression in all plants was confirmed through tissue sampling and enzyme-linked immunosorbent assay (ELISA). WCR eggs from various colonies were incubated at 25°C and 60% rH in total darkness for 13 days until they were about to hatch. At approximately V4, six plants from each maize line were infested with 2,000 WCR eggs per plant. Twenty-four days after infestation, roots were removed from pots and evaluated for larval feeding using the 0 to 3 node injury scale (NIS) (50). Nontransgenic corn (negative control) and current commercial rootworm-protected corn hybrids (SmartStax, MON88017, or Herculex RW) were planted and infected with the same number of rootworm eggs as controls. An NIS score of less than 0.25 is considered a threshold for advancement for further testing.

Field trial root protection assay.Ten locations across the Corn Belt were chosen in 2017 due to high WCR pressure the previous season. Three individually randomized blocks of rows of transgenic lines were planted as replications in each location in late April through mid-May. Usual agronomic practices except application of insecticides were followed for each location. Tassels were removed at VT/T1 stages in mid- to late July. Ten plants from each row were randomly selected and dug for assessment of corn rootworm damage using the 0 to 3 NIS. The highest pressure was observed in the trials from Colesburg, Iowa, and Fairbank, Iowa, where the mean NIS of ≥2.0 on negative controls was observed.

Adult beetle emergence trial.Six replications of isoline and mMpp75Aa1-expressing maize (with relatively low and high expression) were planted in Nebraska at two locations (Leigh and Shelby) with putative high WCR pressure in late April to early May 2017. Normal agronomic practices for the area were followed, and no insecticides were applied at any time. Leaf tissue was collected from each plant at ∼V2 stage and tested with PCR for genes of interest. Plants that did not test as expected were manually removed, and the remaining plants were thinned to approximately the same density. Tents (3 m by 3 m by 2 m) were erected over plants for each maize line prior to the onset of adult beetle emergence. Adults in each tent were collected every 5 to 7 days until no beetles were observed to emerge for 10 days. WCR beetles were identified and counted for each collection. The percent reduction in adult emergence compared to the control was determined as described previously (29).

Larval recovery assay.A single kernel of maize seed from mMpp75Aa1, Herculex RW, or wild-type control maize was planted into 2.36-liter containers (U.S. Plastics, Lima, OH) filled with Berger BM6 potting mix. Each pot had a hole of ∼6.35 mm in diameter with a 25 mm by 25-mm, 530-micron Amber Lumite mesh screen (BioQuip Products, Rancho Dominguez, CA) glued over to facilitate proper drainage. Ten plants per line were maintained in a growth chamber under conditions for growing maize (25°C, 16:8-h light:dark cycle, 50% relative humidity) and watered and fertilized (Peter’s 20-20-20 fertilizer at 300 to 360 ppm nitrogen) as necessary. Each container was infested with 30 neonates from nondiapausing-susceptible (Crop Characteristics) or Cry34Ab1/Cry35Ab1-resistant WCR (8, 51) when plants reached approximately the V4 growth stage. Larvae were left to feed for 10 days, after which root and soil were removed from pots and placed on a Berlese funnel for 3 days. Larvae were collected into 50% ethanol, with counts and instars recorded following collection. Individual larvae were given a numerical value based on mortality (0) or instar stages (1st, 2nd, and 3rd). Adopted from a method described previously (52), a larval instar score (LIS) was used to quantify the average larval development of the 30 neonate larvae infested on each maize plant, calculated as (0 × no. of mortality + 1 × no. of 1st instar + 2 × no. of 2nd instar + 3 × no. of 3rd instar)/30. Between-treatment comparisons were based on the statistical analysis of all observed LIS with the statistical model y_ijk = μ + T_i + C_j + (TC)_ij + e_ijk, where μ is the overall mean across all combinations of maize lines and colonies in all replicates, T_i is the diet treatment effect from maize line i, C_j is the rootworm colony effect from colony j, (TC)_ij is the interaction effect between maize line i and colony j, and e_ijk is the residual effect for replication k. A square root transformation was applied to account for the count nature of the data. Analysis was performed with SAS version 9.4.

Data availability.The sequences for the 3 genes and protein translations are available in GenBank. The accession number for the TIC3670 (Cry75Aa1) gene is MF490291.1, and that for the protein is ASY04853.1. The accession number for the TIC3669 (Cry75Aa2) gene is MF490290.1, and that for the protein is ASY04852.1. The accession number for the TIC3668 (Cry75Aa3) gene is MF490289.1, and that for the protein is ASY04851.1. The protein sequences and new nomenclature are available at the Bacterial Pesticidal Protein Resource Center (BPPRC) (https://www.bpprc.org/). Cry75Aa1 has been renamed Mpp75Aa1, Cry75aa2 has been renamed Mpp75Aa2, and Cry75Aa3 has been renamed Mpp75Aa3.