15 (4): gmr15049099 genes with primers MpCOIF/ MpCOIR, and partial NS and VP sequences with primers MpnDVF1/MpnDVR1.The respective 655-, 1461-, and 423-bp COI, NS, and VP fragments were used to analyze the genetic diversity of MpnDV using MEGA 6.0 and DnaSP 5.0.The high level of identity shared by all COI sequences (>99%) suggested that the aphids sampled were of the same species, and indicated population homogeneity across the 10 locations investigated.The nucleotide diversity of MpnDV sequences (0.0020 ± 0.0025) was significantly higher than that of the COI genes (0.0002 ± 0.0005).The pairwise fixation index for MpnDV was 0.832, and the total gene flow was 0.05.Phylogenetic analysis revealed that the MpnDV haplotypes clustered according to geographical location, except for those from the Liaoning and Shanxi provinces.In conclusion, MpnDV demonstrated a low level of gene flow and high genetic diversity, suggesting that it is vertically transmitted, and implying that endosymbiotic viruses could be used as markers in studies of insect population genetics.
Nicotine is one of the most toxic secondary plant metabolites in nature and it is highly toxic to herbivorous insects. The overexpression of CYP6CY3 and its homologous isozyme CYP6CY4 in Myzus persicae nicotianae is correlated with nicotine tolerance. The expanded (AC)n repeat in promoter is the cis element for CYP6CY3 transcription. These repeat sequences are conserved in the CYP6CY3 gene from Aphis gossypii and the homologous P450 genes in Acyrthosiphon pisum. The potential transcriptional factors that may regulate CYP6CY3 were isolated by DNA pulldown and sequenced in order to investigate the underlying transcriptional regulation mechanism of CYP6CY3. These identified transcriptional factors, AhR and ARNT, whose abundance was highly correlated with an abundance of the CYP6CY3 gene, were validated. RNAi and co-transfection results further confirm that AhR and ARNT play a major role in the transcriptional regulation of the CYP6CY3 gene. When the CYP6CY3 transcript is destabilized by AhR/ARNT RNAi, the transcription of the CYP6CY4 is dramatically up-regulated, indicating a compensatory mechanism between the CYP6CY3 and CYP6CY4 genes. Our present study sheds light on the CYP6CY3 and CYP6CY4 mediated nicotine adaption of M. persicae nicotianae to tobacco. The current studies shed light on the molecular mechanisms that underlie the genotypic and phenotypic changes that are involved in insect host shifts and we conclude that AhR/ARNT regulate the expression of CYP6CY3 and CYP6CY4 cooperatively, conferring the nicotine adaption of M. persicae nicotianae to tobacco.
Recent advances in next generation sequencing (NGS) (e.g. metagenomic and transcriptomic sequencing) have facilitated the discovery of a large number of new insect viruses, but the characterization of these viruses is still in its infancy. Here, we report the discovery, using RNA-seq, of three new partiti-like viruses from African armyworm, Spodoptera exempta (Lepidoptera: Noctuidae), which are all vertically-transmitted transovarially from mother to offspring with high efficiency. Experimental studies show that the viruses reduce their host's growth rate and reproduction, but enhance their resistance to a nucleopolyhedrovirus (NPV). Via microinjection, these partiti-like viruses were transinfected into a novel host, a newly-invasive crop pest in sub-Saharan Africa (SSA), the Fall armyworm, S. frugiperda. This revealed that in this new host, these viruses appear to be deleterious without any detectable benefit; reducing their new host's reproductive rate and increasing their susceptibility to NPV. Thus, the partiti-like viruses appear to be conditional mutualistic symbionts in their normal host, S. exempta, but parasitic in the novel host, S. frugiperda. Transcriptome analysis of S. exempta and S. frugiperda infected, or not, with the partiti-like viruses indicates that the viruses may regulate pathways related to immunity and reproduction. These findings suggest a possible pest management strategy via the artificial host-shift of novel viruses discovered by NGS.
Fall armyworm has invaded China and colonized its populations in tropical and sub-tropical regions of South China since December 2018. Chemical spray has been widely used to control the pest, which shall lead to resistance evolution. In this research, we collected five populations of the pest from Yunnan, Hainan, Tibet, and Fujian of China, and tested their susceptibilities to pyrethroid, organophosphorus, oxadiazine, diamide, antibiotics and other types of insecticides (14 insecticides totally) in the laboratory. Based on the susceptible baseline published from the previous studies, the resistance ratio was 615–1 068-fold to chlorpyrifos, 60–388-fold to spinosad, 26–317-fold to lambda-cyhalothrin, 13–29-fold to malathion, 9–33-fold to fenvalerate, 8–20-fold to deltamethrin, 3–8-fold to emamectin benzoate and 1–2-fold to chlorantraniliprole, respectively. The median lethal concentration (LC50) of other six insecticides without the susceptible baselines was 148.27–220.96 μg mL−1 for beta-cypermethrin, 87.03–128.43 μg mL−1 for chlorfenapyr, 16.35–99.67 μg mL−1 for indoxacarb, 10.55–51.01 μg mL−1 for phoxim, 7.08–8.78 μg mL−1 for M-EBI (the mixed insecticide of emamectin benzoate and indoxcarb) and 1.49–4.64 μg mL−1 for cyantraniliprole. This study can be helpful for chemical control as well as for resistance monitoring and management of the pest in China.
Atmospheric concentrations of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs), polybrominated dibenzo-p-dioxins and dibenzofurans (PBDD/Fs), and polybrominated diphenyl ethers (PBDEs) were measured in Taizhou, a large electronic equipment waste (e-waste) recycling area in East China. The mean concentrations (in summer and winter) of PCDD/Fs (0.45 and 0.39 pg WHO-TEQ m−3, where WHO-TEQ is the toxic equivalent set by the World Health Organisation), PBDD/Fs (0.22 and 0.18 pg WHO-TEQ m−3), and PBDEs (270 and 225 pg m−3) in this region have declined compared with those in 2005, due to regulations on primitive e-waste recycling activities. However, these concentrations remain higher than the historically highest levels in Europe and North America. The congener profiles of 2,3,7,8-substituted PCDD/Fs were similar, with OCDD, 1,2,3,4,6,7,8-HpCDF, OCDF, and 1,2,3,4,6,7,8-HpCDD being the most abundant congeners at all sites. The PCDD/F homologue profiles in the present study were different from those typically observed at non-e-waste locations, indicating a distinct source in this region. Seasonal differences were found in the lower brominated PBDE profiles. These differences indicate that the PBDE emission sources in summer (e.g., strong evaporation sources) differed from those in winter. However, the relatively steady congener profiles of the highly brominated PBDEs suggest that these PBDEs were controlled primarily by similar emission mechanisms. The lifetime excess cancer risks from exposure to PCDD/Fs and PBDD/Fs via inhalation ranged from 0.7 × 10−5 to 5.4 × 10−5, or approximately 80 cancer cases in the Taizhou population.
Article Figures and data Abstract Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract Bacillus thuringiensis (Bt) crops have been widely planted and the effects of Bt-crops on populations of the target and non-target insect pests have been well studied. However, the effects of Bt-crops exposure on microorganisms that interact with crop pests have not previously been quantified. Here, we use laboratory and field data to show that infection of Helicoverpa armigera with a densovirus (HaDV2) is associated with its enhanced growth and tolerance to Bt-cotton. Moreover, field monitoring showed a much higher incidence of cotton bollworm infection with HaDV2 in regions cultivated with Bt-cotton than in regions without it, with the rate of densovirus infection increasing with increasing use of Bt-cotton. RNA-seq suggested tolerance to both baculovirus and Cry1Ac were enhanced via the immune-related pathways. These findings suggest that exposure to Bt-crops has selected for beneficial interactions between the target pest and a mutualistic microorganism that enhances its performance on Bt-crops under field conditions. Introduction Transgenic crops expressing insecticidal Cry proteins from Bacillus thuringiensis bacteria, known as Bt-crops, have become important tools for the management of insect crop pests (Carrière et al., 2003; Cattaneo et al., 2006; Hutchison et al., 2010; Shelton et al., 2002; Tabashnik et al., 2010). Planting of Bt-crops effectively suppresses the targeted insects, decreasing insecticide use and promoting biocontrol services (Bravo et al., 2011; Carrière et al., 2003; Cattaneo et al., 2006; Hutchison et al., 2010; Lu et al., 2010; Lu et al., 2012; Shelton et al., 2002; Tabashnik et al., 2002; Wu, 2010; Wu et al., 2008). We have previously shown that the commercialization of transgenic Bt-cotton in China brought significant changes in the ecology of insects utilizing the crop (Lu et al., 2010; Swiatkiewicz et al., 2014; Wu et al., 2008). However, the effect of Bt-crops on other organisms, such as microbes, which could be playing important roles in the life cycle of insect populations, remains largely unknown. Recently, we showed in laboratory trials that infection with a densovirus, Helicoverpa armigera densovirus-1 (HaDV2, named as HaDNV-1 previously, GenBank accession number: NC_015718), was associated with significantly enhanced tolerance of cotton bollworm, H. armigera, to a baculovirus (H. armigera nucleopolyhedrovirus, HaNPV) (Xu et al., 2014), and there was some suggestion that the densovirus also increased tolerance to Cry1Ac toxin in a Bt-susceptible strain of H. armigera (Xu et al., 2014; Xu et al., 2017a; Xu et al., 2017b). HaDV2 was found to be widespread in wild populations of H. armigera adults (>67% prevalence between 2008 and 2012) (Xu et al., 2014). The densovirus was mainly distributed in the fat body of the insect and could be both horizontally and vertically transmitted. Moreover, HaDV2-positive individuals showed faster development and higher fecundity than non-infected individuals. There was no evidence for a negative effect of HaDV2 infection on H. armigera in relation to other fitness-related traits, suggesting a possible mutualistic interaction between the cotton bollworm and HaDV2 (Xu et al., 2014). Here, we further explore the interaction between H. armigera, HaDV2, and Bt-cotton, to establish its relevance to field populations and to test the hypothesis that the widespread adoption of Bt-cotton in China has selected for cotton bollworm carrying the densovirus. Laboratory experiments show that HaDV2 infection in both Cry1Ac-resistant and Cry1Ac-susceptible strains of cotton bollworm enhances larval tolerance to Bt and overall fitness. Field experiments indicate that H. armigera populations infected with HaDV2 also have a higher tolerance to Bt-cotton relative to those not infected with HaDV2. Moreover, field monitoring over a 10-year period indicates that the frequency of HaDV2-infected cotton bollworm is significantly higher in regions planted with Bt-cotton than in those areas where Bt-cotton is not grown, and that in Bt-cotton-growing areas, the prevalence of HaDV2 infection increases with time since Bt-cotton adoption and with the proportion of cotton that is grown which is transgenic. Further, we found that increased Bt tolerance is associated with activated immune pathways to HaDV2 infection. These results indicate that increased HaDV2 infection in cotton bollworm is correlated with the wide-scale adoption of Bt-cotton in China. Our data are consistent with the notion that exposure to Bt-crops selects for beneficial interactions between the target pest and a microorganism that enhances their fitness in response to Bt exposure under field conditions. Results HaDV2 infection increases Cry1Ac tolerance in H. armigera Previous laboratory bioassays suggested that when a Cry1Ac-susceptible strain of H. armigera was infected with HaDV2, it increased its tolerance to the Cry1Ac toxin (Xu et al., 2014). To explore the generality of this finding, we first analyzed the effect of HaDV2 infection in different H. armigera populations that differ in their susceptibility to Cry1Ac due to different mechanisms of resistance. Two susceptible H. armigera strains infected with HaDV2 (96S and LF) both showed 1.5 times greater tolerance to Cry1Ac toxin, relative to their corresponding non-infected controls (Supplementary file 1a and b). In the case of the Cry1Ac-resistant strains (BtR, 96CAD, LFC2, LF5, LF60, LF120, and LF240), infection with HaDV2 again showed a significant increase in their tolerance relative to the corresponding strains without HaDV2 infection, ranging between 30% and 130% enhanced tolerance (Figure 1). The slope of the regression line is significantly greater than one (t-test: t=2.853, df=6, p=0.029), suggesting that the benefits of carrying HaDV2 may increase with increasing levels of Bt tolerance. Logistic regression confirmed that for a given strain of H. armigera, larvae harboring HaDV2 were significantly more tolerant to Cry1Ac (GLM: Strain: χ28=36.57, p<0.0001, log10 (Cry1Ac toxin concentration): χ21=848.16, p<0.0001; Strain*log10(Cry1Ac toxin concentration): χ28=97.30, p<0.0001; HaDV2-status: χ21=9.53, p=0.0020). However, there was no evidence that the benefits of hosting HaDV2 are affected by the tolerance level of the H. armigera strain, as reflected in the LC50 of the non-infected insects (χ21=0.06, p=0.80). At 8 days post-hatching, H. armigera larvae were significantly lighter (t=10.164, df=32, p<0.0001, n=17) and had lower HaDV2 viral loads in individuals feeding on diet containing Bt than the ones without Bt (t=4.527, df=32, p<0.0001, n=17), suggesting that Bt decreased the replication rate of HaDV2 by suppressing the growth of H. armigera larvae (Figure 1—figure supplement 1). Figure 1 with 1 supplement see all Download asset Open asset Relationship of different Helicoverpaarmigera strains’ LC50 with or without HaDV2 infection. The x-axis is the LC50 of different strains (LF, 96S, LF5, LF60, LF120, LF240, LFC2, 96CAD, and BtR) without HaDV2 infection (HaDV2-negative); the y-axis is the LC50 of different strains (LF, 96S, LF5, LF60, LF120, LF240, LFC2, 96CAD, and BtR) with HaDV2 infection (HaDV2-positive). The regression line is described by the following equation: y=1.2794x+2.8726, R²=0.9606, F=11.99, df=1,7 (p<0.0085). Figure 1—source data 1 Source data for Figure 1. https://cdn.elifesciences.org/articles/66913/elife-66913-fig1-data1-v2.docx Download elife-66913-fig1-data1-v2.docx HaDV2 infection reduces the fitness cost of H. armigera associated with Cry1Ac-resistance evolution To determine whether infection with HaDV2 reduces the costs associated with evolving resistance to Bt, a range of fitness traits (Supplementary file 1c) were measured in four strains of H. armigera that have different Bt-resistance levels (LF, LF5, LF60, and LF240) and were infected or not infected with HaDV2. Enhanced Bt-resistance in H. armigera LF, LF5, LF60, and LF240 strains was associated with lower larval survival rates, prolonged larval and pupal development (in both sexes), reduced pupal weight, lower adult emergence rate, and reduced fecundity and egg hatch rate; in contrast, sex ratio at emergence and the longevity of adults of both sexes were not influenced by the capacity to resist Cry1Ac toxin (Figure 2; Supplementary file 1c and d). When traits were combined to estimate the fundamental net reproductive rates of the four H. armigera strains, R0 (Wang et al., 2016), this revealed that in the absence of HaDV2 infection, the fitness of the most Bt-resistant strain (LF240) was around 40% of that of the most Bt-susceptible strain (LF), consistent with a large fitness cost of resistance (Figure 3); strains with intermediate levels of resistance (LF5 and LF60) suffered a lower cost of resistance (c. 30% reduction in fitness). Figure 2 Download asset Open asset Relationship of different Helicoverpaarmigera strains’ larval (A), and pupal (B,C) with or without HaDV2 infection. (A) The x-axis is the larval development rate (1/duration) of different strains (LF, LF5, LF60, and LF240) without HaDV2 infection (HaDV2-negative); the y-axis is the larval development rate (1/duration) of different strains (LF, LF5, LF60, and LF240) with HaDV2 infection (HaDV2-positive), y=1.0977x−0.0036, R²=0.9921, F=176.678, df=1.2, p=0.006. (B) The x-axis is the female pupal development rate (1/duration) of different strains (LF, LF5, LF60, and LF240) without HaDV2 infection (HaDV2-negative); the y-axis is the female pupa period (1/duration) of different strains (LF, LF5, LF60, and LF240) with HaDV2 infection (HaDV2-positive), y=1.3072x−0.0208, R²=0.9879, F=125.211, df=1.2, p=0.008. (C) The x-axis is the male pupa period (1/duration) of different strains (LF, LF5, LF60, and LF240) without HaDV2 infection (HaDV2-negative); the y-axis is the male pupa period (1/duration) of different strains (LF, LF5, LF60, and LF240) with HaDV2 infection (HaDV2-positive), y=1.0836x−0.003, R²=0.8482, F=7.581, df=1.2, p=0.110. Figure 2—source data 1 Source data for Figure 2. https://cdn.elifesciences.org/articles/66913/elife-66913-fig2-data1-v2.docx Download elife-66913-fig2-data1-v2.docx Figure 3 with 1 supplement see all Download asset Open asset Effects of HaDV2 infection on the net reproductive rate (R0) in four Helicoverpaarmigera strains differing in their tolerance to Bt and not exposed to Cry1Ac toxin. Mean R0 is calculated as the number of female offspring per female that reaches adulthood. The bars are bootstrapped standard errors. Infection with HaDV2 rescued, or partially rescued, this fitness loss, in all four H. armigera strains showing a significant increase in R0 relative to their non-infected counterparts, averaging around 38% higher (paired t-test: t=4.831, df=3, p=0.017; Figure 3). Indeed, when infected with HaDV2, the R0 values of two of the three resistant strains (LF5 and LF60) were comparable to that of the non-infected susceptible (LF) strain (Figure 3). To quantify the fitness cost associated with Cry1Ac-tolerance in the more realistic context of multiple plant defenses, the growth rate of the different strains of H. armigera larvae was also analyzed on Bt-cotton plant leaves. Larval weight after 9 days growth was significantly affected by the cotton variety (Bt or non-Bt) and by the HaDV2 infection-status (Supplementary file 1e and f), with larvae generally being heavier when fed with non-Bt-cotton than with Bt-cotton, and heavier for HaDV2-infected larvae than for larvae not infected with HaDV2; larvae were also heavier when they expressed lower levels of Bt-resistance, indicating that the cost of resistance is reflected in larval growth (Supplementary file 1e and f). None of the interactions between these three main effects explained any additional variation (model comparison with and without interaction terms: F=0.601, df=14, p=0.86), suggesting that the effects of host plant, H. armigera strain, and infection-status on larval growth were additive. HaDV2 infection levels in field populations of H. armigera have increased over the adoption period of Bt-cotton To determine if infection with HaDV2 could increase the performance of H. armigera when exposed to Bt-cotton in the field, we collected H. armigera moths from Xiajin (Shandong Province) and Anci (Hebei Province) in northern China, two locations where Bt-cotton has been widely planted over the last decade (An et al., 2015). Across two successive years, the prevalence of HaDV2 was extremely high (98% of 637 larvae in 2015; 97% of 180 larvae in 2016). Moreover, across both years, the relative average development rates (RADRs) (An et al., 2015) of individuals infected with HaDV2 were significantly higher than that of larvae not infected with the virus (0.62 vs. 0.52 in 2015; 0.61 vs. 0.53 in 2016) (linear model: infection status: F=28.80; df=1.815, p<0.0001; year: F=2.10, df=1.814, p=0.15; infection*year: F=0.57, df=1.813, p=0.57) (Figure 4). Figure 4 Download asset Open asset Frequency distributions for RADR scores for HaDV2 positive and negative insects. The data were collected from field-collected insects from Xiajin and Anci in 2015 and 2016. RADR, relative average development rate. Figure 4—source data 1 Source data for Figure 4. https://cdn.elifesciences.org/articles/66913/elife-66913-fig4-data1-v2.xlsx Download elife-66913-fig4-data1-v2.xlsx Given the apparent selective advantage of HaDV2 infection for insects feeding on Bt-cotton, we predicted that over time we would observe an increase in HaDV2 infection rates in the field and that this would be associated with a temporal increase in average development rates for larvae feeding on Bt-cotton plants. As predicted, over the 10-year period between 2007 and 2016, at both Xiajin and Anci provinces, HaDV2 infection rates increased significantly over time (logistic regression: Xiajin: χ21=405.79, p<0.0001; Anci: χ21=325.21, p<0.0001) (Figure 5A and B). Associated with this, there was a significant temporal increase in larval development rates (RADR) at both locations (linear models: Xiajin: F=5.474, df=1.8, p=0.047; Anci: F=23.256, df=1.8, p=0.0047) (Figure 5C and D). Moreover, across the 10 years at both monitoring locations, there was a strong positive association between HaDV2 infection levels and RADRs, consistent with a possible causal relationship between these two temporal trends (linear models: Xiajin: F=23.826, df=1.9, P=0.001; Anci: F=13.676, df=1.9, P=0.006) (Figure 5E and F). Figure 5 Download asset Open asset HaDV2 infection rate and RADR dynamics and their relationship for each year in the Xiajin and Anci populations during 2007–2016. (A) Relation between HaDV2 infection rate of larvae in Xiajin populations and planting year of Bt-cotton. Logistic regression model of HaDV2 infection rate, logit (y) = 0.49473x−993.1444, R²=0.8591, χ2=405.79, df=1, p<0.0001. (B) Relation between HaDV2 infection rate of larvae in Anci populations and planting year of Bt-cotton. Logistic regression model of HaDV2 infection rate, logit (y) = −0.0105x2+42.1986x−42501.98515, R²=0.877, χ2=325.21, df=1, p<0.0001. (C) Relation between RADR of larvae in Xiajin populations and planting year of Bt-cotton. Linear model of RADR, y=0.009x−17.595, R²=0.406, F=5.474, df=1.8, p=0.047. (D) Relation between RADR of larvae in Anci populations and planting year of Bt-cotton. Linear model of RADR, y=0.016x−32.340, R2=0.744, F=23.256, df=1.8, P=0.001. (E) Relationship of larvae RADR in Xiajin population and HaDV2 infection rate during the years 2007–2016, each data point is a different year, in the Linear model of RADR, y=0.183x+0.438, R²=0.749, F=23.826, df=1.9, p=0.001. (F) Relationship of larvae RADR in Anci populations and HaDV2 infection rate during the years 2007–2016, each data point is a different year, in the Linear model of RADR, y=0.213+0.408, R²=0.625, F=13.676, df=1.9, p=0.006. The bars are the standard error of the mean RADR for the field-derived strains tested in each year. RADR, relative average development rate. Figure 5—source data 1 Source data for Figure 5. https://cdn.elifesciences.org/articles/66913/elife-66913-fig5-data1-v2.docx Download elife-66913-fig5-data1-v2.docx Across regions the HaDV2 infection rates increase with increasing exposure to Bt-cotton To further test the association between H. armigera densovirus infection levels and the adoption of Bt-cotton, we monitored HaDV2 infection rates at 36 locations across 16 provinces during the period 2014–2016, including locations planted with transgenic Bt-cotton (29 monitoring points across 12 provinces) and locations where Bt-cotton has not been planted (9 monitoring points across 4 provinces) (Figure 6; Supplementary file 1g). Across all 3 years, HaDV2 infection levels in H. armigera were significantly higher at locations where Bt-cotton was planted (mean=82%) than in those where it was not (15%) (Figure 6—figure supplement 1) (logistic regression: crop (Bt vs. non-Bt): χ21=354.15, p<0.0001). There was also a significant year-by-crop interaction (χ21=24.13, p<0.0001) due to HaDV2 infection levels being uniformly high across the 3 years at sites growing Bt-cotton (81–90%), whereas infection levels gradually increased from 2014 to 2016 at non-Bt sites (12%, 16%, and 44%, respectively). Figure 6 with 2 supplements see all Download asset Open asset Distribution of HaDV2 in Helicoverpaarmigera from different populations. The red proportion of circles refers to infected individuals, and the blue refers to non-infected individuals. There are significant differences in HaDV2 infection rates between the 29 Bt-cotton planting points and 7 non-Bt-cotton planting points (code: 12, 29, 30, 31, 32, 49, and 50). The sample information was summarized in Supplementary file 1g. Moreover, across the provinces where Bt-cotton is grown, the mean prevalence of HaDV2 in H. armigera between 2014 and 2016 increased with the number of years since Bt-cotton was first introduced (χ21=173.59, p<0.0001) (Figure 7A and Figure 6—figure supplement 1) and also increased as the proportion of cotton that was transgenic increased (χ21=5.34, p=0.021) (Figure 7B). HaDV2 prevalence was not correlated with the proportional area of any of the other crops grown (χ21<1.310, p>0.25). Adding environmental variables to this minimal model, either singly or in combination, did not significantly improve the model fit (average rainfall: χ21=0.110, p=0.74; average temperature: χ21=0.155, p=0.69; average altitude: χ21=0.001, p=0.98; rainfall+temperature+altitude: χ23=0.572, p=0.90). These results are consistent with the notion that the benefits of HaDV2 infection are greatest for H. armigera exposed to Cry1Ac-producing cotton and that selection favoring HaDV2 infection increases the longer the insects are exposed to the Bt-cotton. Figure 7 Download asset Open asset HaDV2 infection rate has increased significantly since Bt-cotton was first introduced and was positively related to the Bt adoption in all cotton areas. (A) Temporal changes in the infection rate of HaDV2 since the introduction of Bt-cotton. (B) Changes in the infection rate of HaDV2 according to the proportion of Bt-cotton in all cotton. Each symbol represents an individual province sampled for densovirus over three years (2014–2016); the mean virus prevalence (± standard error) over those 3 years is shown. Symbol size reflects sampling effort and represents data from >1500 insects. Circles represent the 12 provinces where Bt-cotton is grown; squares are the 4 provinces where Bt-cotton is not grown. The solid line represents the logistic regression (± standard error, shaded zone) describing the relationship between virus prevalence and years since the introduction of Bt-cotton to a province for the 12 Bt-cotton-growing provinces only. The dashed line extrapolates this regression line to Year 0. The detailed information is summarized in supplementary file 1g and Figure 6—figure supplement 1. HaDV2 infection activated the immune pathways in the cotton bollworm To try to understand better the mechanisms increasing the Cry1Ac tolerance levels and enhanced fitness of HaDV2-infected insects, we conducted an RNA sequencing experiment (Supplementary file 1h). The principal component analysis of the transcriptome with differentially expressed genes (DEGs) data clearly distinguished HaDV2-positive from HaDV2-negative individuals at three different time points: 24, 48, and less so at 72 hr after HaDV2 inoculation (Figure 8A,B,C). Taken together with the hierarchical clustering of these DEGs, these results suggest that the HaDV2 has a major effect on the gene expression profiles of their hosts. Figure 8 with 1 supplement see all Download asset Open asset Transcriptome analysis of HaDV2-positive individuals compared to related HaDV2-negative individuals (HaDV2+ vs.HaDV2−) of Helicoverpaarmigera. (A–C) PCA of global gene expression of DEGs at 24 (A), 48 (B), and 72 hr (C) after HaDV2 inoculation. Blue stands for HaDV2-positive samples and red stands for HaDV2-negative samples. (D–F) Heatmaps of –log10 p-values of KEGG pathways representing the upregulated and downregulated DEGs at 24 (D), 48 (E), and 72 hr (F). ‘*’ indicates the significantly enriched pathways (p<0.05). Red color shows upregulation pathways, green color shows downregulation pathways, gray color shows no value, the redder/greener the color, the lower p-values. DEG, differentially expressed gene; PCA, principal component analysis. We performed pathway enrichment analysis on the DEGs, focusing particular attention on pathways related to the development and immune systems (Figure 8D,E,F). Genes in Jak-STAT immune signaling pathway, which are related to some antiviral and antibacterial mechanisms, were significantly enriched and upregulated in the HaDV2-infected larvae at 24 and 48 hr (Figure 8D,E and Figure 8—figure supplement 1A,B), but not at 72 hr (Figure 8F). Interestingly, genes in ABC transporters pathway at 48 hr (Figure 8—figure supplement 1C), the mitogen-activated protein kinase (MAPK) signaling and lysosome pathways at 72 hr (Figure 8—figure supplement 1D,E), which are related to antimicrobial immune response, are significantly enriched and upregulated (Figure 8E,F). Genes in pathways related to development were also significantly enriched, for example, insulin, the mammalian target of rapamycin (mTOR), AMP-activated protein kinase (AMPK) signaling, and the insect hormone biosynthesis pathways at 24 hr (Figure 8—figure supplement 1F,G,H,I), steroid hormone biosynthesis and insulin signaling pathways at 48 hr (Figure 8—figure supplement 1J,K), and the mTOR, protein digestion and absorption, and steroid hormone biosynthesis pathways at 72 hr. These results may help to explain why HaDV2-positive individuals developed more quickly than non-infected insects. HaDV2 decreased the effect of Bt on H. armigera There was a total of 1573 significant DEGs in HaDV2-negative insects after exposure to Cry1Ac (673 upregulated and 900 downregulated). We focused on DEGs and pathways related to Bt tolerance and immune systems. Seven ABC transporter genes, which are related to immunity, were differentially expressed (four upregulated and three downregulated) (Figure 9A); eight trypsin genes, which are related to the conversion of the protoxin to activated toxin, were differently expressed (six upregulated and two downregulated) (Figure 9B); two carboxylesterase genes, which are related to the detoxification of Bt by the insect, were upregulated; and two Bt toxin receptors genes, alkaline phosphatase (ALP) and aminopeptidase N (APN), which are related to Bt resistance, were downregulated (Figure 9C). Genes in the MAPK signaling pathway, which is related to antimicrobial immune response, was also significantly downregulated (Figure 9D). Figure 9 Download asset Open asset Transcriptome analysis of Helicoverpa armigera after HaDV2 infection and Cry1Ac exposure. The quantity of DEGs with log2(FPKM) related to the expression of (A) the ABC transporters; (B) trypsin; (C) Bt receptors and carboxylesterase genes; (D) the MAPK signaling pathway; (E) the drug metabolism pathways. DV−=HaDV2-negative individuals, DV+=HaDV2-positive individuals. Bt−=larvae were fed on the artificial diet without Cry1Ac, Bt+=larvae were fed on the artificial diet containing 1 µg/mL Cry1Ac. Colors in log2(FPKM) indicate the gene expression levels, the hotter (redder) the color, the higher the gene expression level. The three columns represent three biological replicates. DEG, differentially expressed gene; MAPK, mitogen-activated protein kinase. In contrast, there were only 249 significant DEGs in HaDV2-positive insects after exposure to Cry1Ac (165 upregulated and 84 downregulated). One trypsin gene was downregulated; genes in the ascorbate and aldarate metabolism pathway, which is related to carbohydrate metabolism were upregulated; genes in drug metabolism – cytochrome P450 pathway, metabolism of xenobiotics by cytochrome P450 pathway and drug metabolism – and other enzymes pathways, which are related to detoxification, were also upregulated (Figure 9E). Discussion Commercialization of transgenic Bt-crops has brought some significant benefits to farmers (Carrière et al., 2003; Cattaneo et al., 2006; Lu et al., 2012; Shelton et al., 2002; Wu et al., 2008). These Bt-plants successfully control target insect pests and have resulted in the reduced use of insecticides in the field (Bravo et al., 2011; Lu et al., 2012; Wu et al., 2008). Here, we show that HaDV2 infection enhances tolerance to Cry1Ac in both susceptible and resistant strains of H. armigera. We also show that there appears to have been a strong positive selection for HaDV2-infected H. armigera in populations exposed to Cry1Ac-cotton in the field, especially (but not exclusively) when this exposure has been for prolonged periods of time. Intensive monitoring in two provinces where Bt-cotton has been grown over a number of years showed a strong temporal increase in HaDV2 infection levels, from around 40% in 2007 to nearly 100% in 2016, with high levels of HaDV2 infection being associated with higher larval growth rates on Bt-cotton in the field. Indeed, at locations where Bt-cotton is not grown, H. armigera shows much lower rates of HaDV2 infection (12–44% vs. 81–90%), consistent with weaker selection for HaDV2-infected H. armigera in the absence of Bt-crops. Moreover, the prevalence of HaDV2 infection increased with the number of years since Bt-cotton was first introduced to a province and with the proportion of cotton grown that is transgenic, consistent with Bt-cotton being a key selection pressure. There were no significant relationships between HaDV2 prevalence and the proportional cover of any other main crops grown locally (cotton, rice, corn, wheat, beans, tubers, oil-producing crops, and vegetables) or with any of the environmental variables tested (rainfall, temperature, and altitude). Thus, our data suggest that the Cry1Ac-toxin selected for HaDV2-infected H. armigera following exposure to Bt-cotton and our RNA-Seq analysis suggests that the increased tolerance of HaDV2-infected insects is caused, in part at least, by the activation of a series of immune pathways and pathways that enhance development. The emergence and rapid spread of a symbiont through a host population have been observed previously (Himler et al., 2011; Turelli and Hoffmann, 1991; Weeks et al., 2007). For example, the bacterial symbiont, Rickettsia sp. nr. bellii, swept through a population of invasive sweet potato whitefly, Bemisia tabaci, in just 6 years (Himler et al., 2011). Although we do not know when HaDV2 first infected H. armigera in China, in two areas where Bt-cotton is grown extensively, and where nearly 100% of all insects are currently infected with HaDV2, the prevalence of the densovirus was as low as 40% in 2007, when the earliest samples were collected (Figure 5). Moreover, extrapolation of the observed temporal trends (Figure 5A,B and Figure 7) would suggest that emergence of HaDV2 may have coincided with the introduction of Bt-cotton in China two decades ago. This apparently rapid spread is perhaps not surprising given the strong fitness benefits of carrying the densovirus (amounting to a ~40% increase in H. armigera R0) (Figure 3). Consistent with this, based on these estimated increases in R0, a simple population growth model (Himler et al., 2011) suggests that HaDV2 would go to near fixation in 20–30 generations in Bt-cotton-growing areas (Figure 3—figure supplement 1). Rather than being a recent introduction to H. armigera in China, it is possible that the association between the HaDV2 and its host is more ancient, but that the increase in HaDV2 prevalence in recent years is because it has only recently evolved from a parasitic or commensal relationship into one that is more mutualistic (Weeks et al., 2007), perhaps under direct or indirect selection from Bt-cotton. Indeed, most of the densoviruses that have been studied thus far have tended to be at the parasitic end of the symbiotic spectrum and some of these have even been promoted as potential biological pesticides (El-Far et al., 2012). However, the observation that densoviruses tend to be parasitic in nature is likely biased by the fact that viruses with notable pathological effects on their hosts are more likely to be reported (Roossinck, 2011; Roossinck, 2015; Webster et al., 2015; Xu et al., 2020). W
The complete genome of a novel virus from Arma chinensis was determined by RNA sequencing and rapid amplification of cDNA ends. This virus has a single-stranded RNA genome of 10,540 nucleotides (nt) excluding the poly(A) tail. Two non-overlapping open reading frames (ORFs) in the sense direction were predicted: one long ORF at the 5' end of the genome (6,219 nt) that encodes a polypeptide of 2,072 amino acids (aa), and one short ORF at the 3' end of the genome (3,033 nt) that encodes a polypeptide of 1,010 aa. Phylogenetic analysis indicated that the virus clusters within a large cluster of currently unidentified picorna-like viruses with a high bootstrap value. We named the virus isolate Arma chinensis picorna-like virus 1 (AcPV-1). The prevalence of AcPV-1 infection in samples of Arma chinensis from the wild was at a low level (5.48%, 8 positives in 146 samples). Keywords: Arma chinensis; genomic characterization; phylogenetic analysis; Arma chinensis picorna-like virus 1; prevalence.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)