Insect rearing
The B. dorsalis strain collected from a carambola (Averrhoa carambola) orchard in Guangzhou, Guangdong Province, was reared under laboratory conditions (27 ± 1 °C, 12:12 h light:dark cycle, 70–80% RH). A maize-based artificial diet containing 150 g of corn flour, 150 g of banana, 0.6 g of sodium benzoate, 30 g of yeast, 30 g of sucrose, 30 g of paper towel, 1.2 mL of hydrochloric acid and 300 mL of water was used to feed the larvae. Adults were fed a solid diet (consisting of 50 g yeast and 50 g sugar) and 50 mL sterile water in a 35 cm × 35 cm × 35 cm wooden cage. For B. dorsalis, the female will start laying eggs once mated and the female will start mating 7 days after emergence. To make sure all females used in our study were gravid females, females were selected 10 day after emergence.
Visualization of CF-BD with FISH and PCR
FISH was carried out on dissected gut and ovary samples from B. dorsalis. The hybridization protocol for the gut and ovary was performed according to a previously described method32. Briefly, the gut and ovary were collected and immediately soaked in Carnoy’s fixative for 12 h. After sample fixation, proteinase K (2 mg/mL) treatment for 20 min at 37 °C and HCl (0.2 mol/L) treatment for 15 min at room temperature were performed successively. Then, followed by dehydration in ethanol, the samples were incubated in buffer (20 mM Tris-HCl (pH 8.0), 0.9 M NaCl, 0.01% sodium dodecyl sulfate, 30% formamide) containing 50 nM CF-BD specific probe (5′-AATGGCGTACACAAAGAG-3′) labeled with cy3 at the 5′ end for 90 min. After incubation, the samples were washed with buffer (0.1 M NaCl, 20 mM Tris/HCl (pH 8.0), 5 mM ethylenediaminetetraacetic acid (pH 8.0), 0.01% SDS) and observed under an epifluorescence microscope (Axiophot, Carl Zeiss, Shinjuku-ku, Japan).
To further confirm CF-BD in rectum and ovary of mature females, rectums and ovaries of mature females were dissected and fixed in formalin fixation for 24 h. After soaking in graded alcohols and xylene, all samples were embedded in paraffin for section preparation. Samples were sliced into 4 µm each before pasting on the glass slide and then sent for FISH with the same probe (labeled with cy3 at the 5′ end) used above. Moreover, nested PCR was applied to detect CF-BD in 19 ovaries of mature females according to the method of Guo et al., 201733. Briefly, a 1149 bp region of gyrB gene of CF-BD was amplified by the specific outer primer gyrBP1-F (5′-CAGCCCACTCTGAACTGTAT-3′) and gyrBP1-R (5′-TCAGGGCGTTTTCTTCGATA-3′) under a temperature profile of 95 °C for 1 min, which was followed by 25 cycles of 95 °C for 30 s, 52 °C for 30 s, 72 °C for 90 s, and 72 °C for 5 min. Then, a 371 bp region of the gyrB gene of CF-BD was amplified by the specific inner primer gyrBP4-F (5′-ACGCTGGCTGAAGACTGCC-3′) and gyrBP4-R (5′-TGGATAGCGAGACCACGACG-3′) under a temperature profile of 95 °C for 2 min, which was followed by 35 cycles of 95 °C for 30 s, 57 °C for 30 s, 72 °C for 30 s, and 72 °C for 5 min.
Influence of CF-BD on B. dorsalis ovary development
To evaluate the effect of CF-BD on ovary development, newly emerged B. dorsalis females were injected with streptomycin and CF-BD suspension (both dilute in sterile water). Specifically, 10 µL 25% glycerol solution containing CF-BD was added into 100 mL Luria-Bertani (LB) liquid medium and culturing for 1 day by shaking (180 rpm) in 30 °C incubator. After culturing, CF-BD was collected by centrifuging (3000 rpm, 15 min) the medium in a 50 mL centrifuge tube. Then collected CF-BD was re-suspended with 5 mL sterile water. CF-BD concentration was measured on a hemocytometer and CF-BD concentrations used in the following assays were prepared by diluting the original concentration with sterile water. A 0.5 mm inside diameter capillary needle with 1 μL streptomycin or CF-BD suspension was used for injection. The injection operation was carried out on a microinjector (Eppendorf FemtoJet), and every female was injected in the abdomen near the ovipositor. The concentrations of streptomycin used were 20 mg/mL, 10 mg/mL and 5 mg/mL, respectively. And CF-BD suspension concentrations were 3 × 107 cfu/mL, 1.5 × 107 cfu/mL and 7.5 × 106 cfu/mL, respectively. For control, the female fly was injected with 1 μL sterile water in the abdomen near the ovipositor. Then the development level of the ovary was assessed by comparing the width and length of ovary between streptomycin (or CF-BD suspension) injection flies and control. For CF-BD injected flies, developmental facilitation was observed for ovaries 2 days before the flies reached sexual maturity (flies will reach sexual maturity after 7 days). For antibiotic injected flies, ovaries were dissected after 7 days.
Oviposition assays
The method reported in previous studies was followed for the oviposition experiments17. Briefly, a 2-choice apparatus was assembled in a cage made up of wood and wire gauze (length: width: height = 60 cm: 60 cm: 60 cm) with two petri dishes (diameter: 3 cm) at the bottom of the cage (Fig. 2a). All devices were sterilized before each experiment. Fresh fruits of guava (Psidium guajava Linn.) and mango (Mangifera indica L.) were sourced from the local market in Guangzhou, China. These fruits were sterilized on the surface with ethanol and ground into puree with a sterilized grinder, and puree (2 g) was added to the sterilized Petri dishes of the cages (one dish with puree containing 100 μL CF-BD (0.8*108 cfu/mL) in sterile water, and one dish with puree containing 100 μL sterile water). Then the prepared cages were divided into two groups for different assays. Group 1: At 0 h, 50 gravid females of B. dorsalis were placed in the cages and egg numbers in the petri dishes were recorded after 2 h. Group 2: At 4 h, 50 gravid females of B. dorsalis were placed in the cages and egg numbers in the petri dishes were recorded after 2 h.
To test the oviposition attraction of 3-HA, a 4-choice apparatus was assembled in a cage made up of wood and wire gauze (length: width: height = 60 cm: 60 cm: 60 cm) with four petri dishes (diameter: 3 cm) at the bottom of the cage. In the Petri dishes, 2 g puree, 2 g puree + 0.2 mg 3-HA, 2 g puree + 2 mg 3-HA and 2 g puree + 20 mg 3-HA were added. Then, the egg-laying behavior was observed31.
To test the oviposition attraction of 3-HA to flies with genes knocked down, 20 females injected with dsRNA were placed into the above cage with two Petri dishes. In the Petri dishes, 2 g guava puree and 2 g guava puree + 20 mg 3-HA were added. Then, the egg-laying behavior was observed using the above method. Oviposition of normally reared females was performed as a control. The oviposition index was calculated using the following formula:
Oviposition index = (O − C)/(O + C), where O is the number of eggs in the treatment and C is the number of eggs in the control.
Volatile analysis
The volatile compounds in guava and mango purees were analyzed by GC–MS according to the method described in a previous study17. Briefly, 2 g puree mixed with sterile water or CF-BD was added into a 20 ml bottle, and then a 100-μm polydimethylsiloxane (PDMS) SPME fiber (Supelco) was used to extract the headspace volatiles for 30 min. GC–MS was performed with an Agilent 7890B Series GC system coupled to a quadruple-type-mass-selective detector (Agilent 5977B; transfer line 250 °C, source 230 °C, ionization potential 70 eV). The 3-HA concentrations in puree mixed with sterile water and CF-BD were measured with the standard curve drawn by the authentic standards of 3-HA. And 3-HA concentration in puree mixed with sterile water and CF-BD was compared with a paired sample Student’s t-test.
Olfactometer bioassays
An olfactometer consisting of a Y-shaped glass tube with a main arm (20 cm length*5 cm diameter) and two lateral arms (20 cm length, 5 cm diameter) was used. The lateral arms were connected to glass chambers (20 cm diameter, 45 cm height) in which the odor sources were placed. To ensure a supply of odor-free air, both arms of the olfactometer received charcoal-purified and humidified air at a rate of 1.3 L/min.
To test the attraction effect of puree supplemented with CF-BD or 3-HA for females, puree mixed with CF-BD was prepared and placed in one odor glass chamber. In the control odor glass chamber, puree mixed with sterile water was placed. After 4 h, gravid females were individually released at the base of the olfactometer and allowed 5 min to show a selective response. The response was recorded when a female moved >3 cm into one arm and stayed for >1 min. Females that did not leave the base of the olfactometer were recorded as nonresponders. Only females that responded were included in the data analysis. Odor sources were randomly placed in one arm or the other at the beginning of the bioassay, and the experiment was repeated ten times. The system was washed with ethanol after every experiment. More than 100 females were selected for testing, and each female was used only once for each odor. A chi-square test was performed to compare the attraction difference between puree mixed with sterile water and CF-BD.
Olfactory trap assays
The attraction of purees supplemented with CF-BD to mature females was also tested. The test chamber was assembled with a plastic cylinder (120 × 30 cm) covered by a ventilated lid. The test chamber contained an odor-baited trap (2 g puree + 100 μL CF-BD (0.8*108 cfu/mL)) and a control trap (2 g puree + 100 μL sterile water). The traps were made of transparent plastic vials (20 × 6 cm) and were sealed with a yellow lid on which small entrances were present to let the flies in (Fig. 3a). After 0 h or 4 h of fermentation, 100 gravid females were released in the cage. The fly number in each trap bottle was recorded after 2 h. The number of flies was compared with a paired sample Student’s t-test.
The attraction effect of puree supplemented with 3-HA on mature females was tested by placing four traps (2 g puree, 2 g puree + 0.2 mg 3-HA, 2 g puree + 2 mg 3-HA and 2 g puree + 20 mg 3-HA) in the test chamber. Then, the attraction effect was observed31.
Video observation of egg-laying behavior
Egg-laying behavior was observed in a Petri dish. Briefly, guava puree was added to a centrifuge tube on which a hole was made. Then, one gravid female was placed into the petri dish, and the lid was closed. Above the petri dish, a camera was placed to record the behavior of the female before laying eggs.
EAG analysis
EAG analysis was performed to determine whether 3-HA could elicit electrogram responses in the ovipositors of gravid females and Obps knocked down gravid females. For EAG preparations, the ovipositor of a gravid female was cut off and mounted between two glass electrodes (one electrode connected with the ovipositor tip). The ovipositor tip was cut slightly to facilitate electrical contact. Dilution of 3-HA in ethanol (0.1, 1 and 10 mg/mL) was used as a stimulant. Ethanol was used as control. For each ovipositor, ethanol and 3-HA diluted in ethanol were used as stimulants. The signals from the ovipositors were analyzed with GC-EAD 2014 software (version 4.6, Syntech).
Transcriptome sequencing and gene identification
To identify the olfactory genes that contribute to B. dorsalis oviposition preference, the transcriptome sequencing results of the female ovipositors at different developmental times (0 day, 3 days, 6 days, 9 days and 12 days) were compared. For each time, 5 ovipositors were dissected for RNA extraction. In addition, five replicates were included for each time. In the next step, paired-end RNA-seq libraries were prepared by following Illumina’s library construction protocol. The libraries were sequenced on an Illumina HiSeq2000 platform (Illumina, USA). FASTQ files of raw reads were produced and sorted by barcodes for further analysis. Prior to assembly, paired-end raw reads (uploaded to National Genomics Data Center, Accession number: PRJCA004790) from each cDNA library were processed to remove adapters, low-quality sequences (Q < 20), and reads contaminated with microbes. The clean reads were de novo assembled to produce contigs. An index of the reference genome of B. dorsalis was built, and paired-end clean reads were mapped to the reference genome using HISAT2. 2.4 with “-rna-strandness RF” and other parameters set as a default68. To evaluate transcript expression abundances, StringTie software was applied for calculation of the normalized gene expression value FPKM69. Then, gene differential expression analysis was performed with DESeq2 software70. Genes/transcripts with a false discovery rate (FDR) below 0.05 and absolute fold change ≥2 were considered differentially expressed genes/transcripts. According to the above results, we screened the differentially expressed olfactory genes.
Trend analysis
Gene expression pattern analysis was used to cluster genes with similar expression patterns for the ovipositor samples. To examine the expression pattern of the differentially expressed genes, the expression data of each sample were normalized and then clustered with Short Time-series Expression Miner software (STEM)71. The clustered profiles with p < 0.05 were considered significant profiles.
Expression validation of olfactory genes
qRT-PCR analysis was used to validate olfactory gene expression in ovipositors at different developmental times. Antennae of 12-day-old female were used as controls. Total RNA from the antennae of 12-day-old females and ovipositors at 0 d, 3 d, 6 d, 9 d and 12 d was extracted. Then, cDNA was synthesized with a One-Step gDNA Removal and cDNA Synthesis SuperMix Kit (TransGen Biotech, Beijing, China) using the extracted RNA. Then, a PerfectStarTM Green qPCR SuperMix Kit (TransGen Biotech, Beijing, China) was used to perform quantitative real-time PCR to compare the gene expression levels. Gene-specific primers (Supplementary Table 1) were designed on NCBI with primer blast. The α-tubulin and actin genes were used as reference genes72. The PCR procedure was set according to the manufacturers’ instruction. Three biological duplicates and 3 technical duplicates were performed.
The ability of the Obps to bind 3-HA
Recombinant proteins Obp56d and Obp56d-2 were obtained by in vitro expression in Escherichia coli Rosetta (DE3) cells mainly according to the previous study73. Briefly, PCR primers (Supplementary Table 2) with restriction sites were designed according to the CDS sequences of Obp56d and Obp56d-2. Then PCR products were purified and connected to the peT-32A (+) Prokaryotic expression vector. The vector was then transformed into Escherichia coli Rosetta (DE3) for expression. The positive clones were screened by kanamycin resistance and sequenced to confirm the correct sequence. The verified clones were cultured in the kanamycin resistance LB medium at 37 °C for 16 h. Then 100 ml bacterial solution was inoculated in the LB medium with the final concentration of 0.1 mM IPTG for 8 h at 18 °C. Then bacteria were collected by centrifugation at 8000 rpm and then re-suspended in lysis buffer (80 mM Tris-HCl, 200 mM NaCl, 1 mM EDTA, 4% glycerol, pH 7.2, 0.5 mM PMSF). Sonication (3 s, five passes) was conducted to break the bacteria cells. The recombinant proteins in the supernatant were collected by centrifugation. Then the proteins were purified by two rounds of anion-exchange chromatography and concentrated by the ultrafiltration cube. Both purified recombinant protein was detected using SDS-PAGE.
The combination fluorescence experiment of Obp with N-phenyl-1-naphthylamine (1-NPN) was performed on a Microplate Reader (Thermo Scientific Varioskan LUX) referring the previous study in locust74. The excitation wavelength was set as 337 nm, and the emission wavelength was 380-520 nm. Obp (2 μM dissolved in 50 mM Tris-HCl, PH = 7.4) and 1-NPN (2μM–16μM dissolved in chromatographic methanol) were mixed and the strongest fluorescence values was recorded. The dissociation constant value of 1-NPN combined with Obp was calculated using one site-specific binding method in GraphPad 8.0. Then the dissociation constants of the Obp56d and Obp56d-2 were generated. For competitive binding assays, 3-HA (0, 4, 8, 12, 16 μM dissolved in chromatographic methanol) was used as competitor to bind the Obp in the Obp/1-NPN complex.
RNA interference of the olfactory genes
Double-stranded RNA (dsRNA) primers (Supplementary Table 2) tailed with the T7 promoter sequence were designed using the CDSs of Obp56d, Obp56d-2, OR7a-2 and IR25a as templates. A MEGAscript RNAi Kit (Thermo Fisher Scientific, United States) was used to synthesize and purify dsRNA according to the manufacturer’s instructions. The GFP gene was set as the RNAi negative control. To knockdown the target gene in females, 0.5 μL (500 ng/μl) dsRNA was injected into the abdomen of an 11-day-old female. Flies injected with dsGFP were prepared as a negative control. After 24 h, the knockdown efficiency of the genes was checked with qRT-PCR following the method used for validating the expression of the olfactory genes above. Then, the oviposition preference for 3-HA was tested in flies in which the olfactory gene was silenced.
Statistics and reproducibility
In box-and-whisker plots, where the boxes encompass the first to the third quartiles, inside the box the horizontal line shows the median, and the whiskers are the maximum and minimum observation. In bar plots, the data is presented as mean ± SEM. ANOVA followed by Tukey’s test, Wilcoxon matched-pairs signed rank test, paired sample student t-test, Kendall nonparametric test, chi-square test and Kruskal–Wallis test followed by Dunn’s test were used to compare difference between experimental treatments. In ANOVA, Kruskal–Wallis test is used instead for the data don’t conform to ANOVA assumptions. Prism 8 (GraphPad, San Diego, CA) was used to do statistical analysis and generate the figures. P ≤ 0.05 were considered as statistically significant. Sample sizes are in all cases stated in the manuscript.
Reporting summary
Further information on research design is available in the Nature Research Reporting Summary linked to this article.
Source: Ecology - nature.com
