Removal or component reversal of local geomagnetic field affects foraging orientation preference in migratory insect brown planthopper Nilaparvata lugens

Insect stock

Experiments were performed at Beijing Key Laboratory of Bioelectromagnetics, Institute of Electrical Engineering, Chinese Academy of Sciences, Beijing, China. The insects were reared in climate chambers at day/night temperatures of (27 ± 1) °C/(26 ± 1) °C on susceptible Taichuang Native 1 (TN1) rice plant under 14:10 h light: dark cycle and 70 ± 5% humidity (Wan et al., 2015) and the environmental conditions of the chambers was the same in the entire experiment. The TN1 rice plants were prepared in advance, and used as the food for the insects when they grew up to 10 cm height. The migratory macropterous female and male adults were selected from the same generation for the successive generations (Wan et al., 2016).

Magnetic field devices setup

The GMF used in the experiment (total intensity: 52,487 ± 841 nT; declination 5.30 ± 0.59°; inclination 56.29 ± 1.02°) were the local GMF at (39°59′14″N, 116°19′21″E). The artificial magnetic fields were produced using a Helmholtz coil system (Fig. 1). For NZMF, the Helmholtz coils were used to produce a near-zero magnetic field region with an average intensity of ∼500 nT at a center spherical space (150 mm in radius). For component reversal of GMF, the Helmholtz coils were used to generate a magnetic field with twice intensity and reversed direction to offset either the horizontal component or the vertical component of GMF, producing a reversed inclination with the same intensity but reversed component of GMF. Routinely before and after each experiment, we measured the three components of GMF using a fluxgate magnetometer (Model 191A, Honor Top Magnetoelectric Technology Co. Ltd., Qingdao, China; sensitivity: ±1 nT) to modulate the electric current of the coil pairs to produce the required intensity for NZMF.

Cross-tube choice system and foraging orientation experiments

The choice system consisted of a cross tube which was embedded in a plastic square. The length of each arm of the cross tube was 110 mm. The width was 20 mm and the depth was 30 mm. The cross tube was covered with a plastic lid of same size. There were small holes at each arm end for air flowing through. A lamp (15 W, λ = 320–680 nm) was installed as the light source (there is faint light when the N. lugens takes off at the sunset or sunrise) with 400 lumen of lumination intensity at the cross tube. The coil system was covered by a shade cloth during the experiment to shield the external environment (Fig. 1). During the experiment, the cross tube was placed horizontally inside the Helmholtz coils and the arms of the cross tube were oriented towards four cardinal points. The cardinal points used in the experiments were the same. Two cross tubes were used in the experiment, one containing food as a reward and the other without food. Ten fresh rice seedlings of susceptible variety of TN1 (Fu et al., 2001) was placed at one arm end as food reward.

The experiment was conducted in two parts: the first part of the trial was to provide food as a stimulus that the insects might associate with magnetic cues under the normal GMF, and the second part of the trial was to test for a disruption of their ability to exhibit this learned directional preference when the GMF was altered. Adult insects within 48 h of emergence regardless of gender or mating status were collected from the rearing colony and introduced into the center of the cross tube using a self-made insect suction-implement. The rice seedings were placed 70 mm away from the center (Fig. 2A). For the first part of the trial, the insects gathering around the rice seedlings (40 ± 13 insects) were eventually collected (Fig. 2B) using the same suction-implement into a vial and afterwards a new cross tube with no rice seedlings inside was placed horizontally in the Helmholtz coils. For the second part of the trial, the collected insects in the vial were replaced in the center of the new cross tube using the suction-implement (Fig. 2C) and the insects moving to each of the four arms of the new cross tube were recorded for 0.5 h (Fig. 2D). The entire experiment was performed at room temperature (26 ± 1 °C) and each magnetic field setup was performed individually for 12 replicates and the total number of insects tested was 511 ± 69.

The effects of IscA1 gene silencing on the orientation of N. lugens

The IscA1 gene was previously cloned in Nilaparvata lugens and the results showed that the gene expression reached the peak at the third day after emergence (Xu et al., 2017), so adults of 1^st day after emergence were selected for RNAi according to Liu et al. (2010). The primers (Table 1) were designed based on the fragment sequence that was searched from transcriptome of N. lugens by local BLAST search. The dsRNA of IscA1 gene was designed at two different regions: nearing the 3′end (dsRNA1) and nearing the 5′end (dsRNA2). Insects were anesthetized with CO₂ for 30 s at PCO₂ = 1 mPa and immobilized on a 1.5% agarose plate with abdomen upward. Each insect was injected with 250 ng (50 nl in volume) dsRNA. On the second day after injection (24–48 h), the injected insects were collected and placed inside the GMF for foraging orientation test with the cross-tube system. The cross-tube behavioral trials for the RNAi insects were conducted in the same manner as described in section 2.3 as part of a two-step trial. A total number of 700 insects were used for the experiments. To ensure the silencing efficiency, the expression level of IscA1 gene was investigated before and after the behavioral test using three pools of six insects for each group by fluorescence-based quantitative real-time PCR (q-PCR) (Bustin et al., 2009). The whole body of adult N. lugens was used for sampling and all the samples were collected during the same time period (19:00–20:00 h). Total mRNA was extracted by TRIzol reagent (Invitrogen, Waltham, MA, USA). The quality of samples was determined by spectrophotometric optical density (OD) 260/280 and 2% agarose gel electrophoresis. The cDNA templates were synthesized with 1 μg of total RNA using PrimeScript™ RT reagent Kit with gDNA Eraser (TaKaRa, Tokyo, Japan). Each cDNA product was diluted with sterilized double- distilled water. The house-keeping gene for the q-PCR was actin1 (GenBank accession No. EU179846), and the PCR amplification efficiency was established by means of calibration curves (Bustin et al., 2009). The optimized thermal program was designed according to the kit instructions. Quantification of the transcript level of genes was conducted according to the ^∆∆Cq method (Livak and Schmittgen, 2001). RNA samples were analyzed independently for three times. The dsRNA of green fluorescent protein-GFP (GenBank accession No. U76561) was injected into the N. lugens as the control.

Statistical analysis

All data were analyzed using SPSS 20.0 (IBM Inc., Armonk, NY, USA.). The chi-square test was used to analyze the ratio of the distribution of insects in four directions. If there was significant difference, Bonferroni correction was used to analyzed the difference between every two directions. One-way ANOVA was used to analyze the gene expression. Significant differences between dsGFP (control) and dsNl-IscA1 injection treatments were determined by one-way ANOVA at p < 0.05.

The foraging orientation preference of N. lugens in the GMF vs NZMF

In the GMF, N. lugens showed the highest preference of foraging orienting to the north direction with original food (χ² = 108.48, p < 0.001). The percentage of individuals orienting to the north was 39.06%, which was significantly higher than that to south 14% (χ² = 98.481, p < 0.001), west 25.40% (χ² = 26.169, p < 0.001) and east 21.39% (χ² = 45.652, p < 0.001, Fig. 3A). In the NZMF, however, N. lugens were relatively equally distributed and the percentage of individuals orienting to the north, south, west and east direction was 25.46%, 22.22%, 27.55% and 24.77%, respectively (χ² = 10.261, p = 0.088) (Fig. 3B).

The foraging orientation preference of N. lugens in the horizontal or vertical component reversal of GMF

In order to study how the GMF affects the foraging orientation ability of N. lugens, we conducted behavior experiment in the horizontal or vertical component reversal of GMF. In the horizontal component reversal of GMF, most of N. lugens were distributed in the north (χ² = 87.872, p < 0.001, Fig. 4A), similar to that in GMF. In the vertical component reversal of GMF, the percentage of insects orienting to the north, south, west and east direction observed as 25.18%, 21.74%, 29.53% and 23.55%, respectively (χ² = 9.371, p = 0.102) (Fig. 4B).

The IscA1 gene knockdown affected the foraging orientation preference of N. lugens

The q-PCR results showed that the mRNA expression of IscA1 was effectively downregulated after the gene silencing. Both of the silencing efficiencies of dsRNA1 and dsRNA2 were over 80% within 24 h (before the behavioral experiment) and 48 h (after the behavioral experiment) after microinjection (Fig. 5). Since the N. lugens preferred the north foraging direction in GMF or horizontal component reversal of GMF, we chose these conditions to investigate whether IscA1 gene silencing would affect the insects’ choice of direction. In these two magnetic fields, most of the N. lugens with IscA1 gene silencing distributed in the west, followed by the north, south and east (Fig. 6). Compared with the wild type, the percentage of IscA1 gene knockdown N. lugens distributed in the north direction revealed significantly decreased from 39.06% to 28.82% in GMF (χ² = 13.183, p < 0.001, Fig. 6A) and to 28.57% in the horizontal component reversal of GMF (χ² = 10.151, p < 0.001, Fig. 6B).