【作者】 刘永杰；

【导师】 冯明光；

【作者基本信息】 浙江大学 ， 微生物学， 2013， 博士

【摘要】 基于活体细胞的真菌杀虫菌剂时效性较差,限制其商业开发和应用。生防真菌转基因工程菌剂在很大程度能克服杀虫时效性较差的缺点,但绕不开环境安全性的公众疑虑,而目前缺乏转基因真菌杀虫剂田间杀虫效果和安全性评价的实例。本研究以先前构建的超高表达苏云金芽孢杆菌(Bacillus thuringiensis)杀虫蛋白Vip3Aal而对鳞翅目害虫表现高胃毒杀虫活性的球孢白僵菌(Beauveria bassiana)工程菌株BbHV8及其亲本野生株Bb2860(简称WT)为材料,将其高纯度分生孢子粉配制成孢子悬乳剂(emulsifiable formulation),在浙江金华从初夏开始进行了两个独立的各历时35天防治以菜青虫(Pieris rapae)为主、小菜蛾(Plutella xylostella)和桃蚜(Myzus persicae)为辅的田间试验,一是比较两种菌剂与当家化学杀虫剂甲氨基阿维菌素苯甲酸盐(简称甲维盐)在根据虫情四次用菌用药条件下对卷心甘蓝的整季保护效果；二是在试验期间评价了各处理对田间捕食性天敌蜘蛛种群的影响,试验之后数月期间多批次采样评价两种菌剂在释放环境中的存留状况、环境适应性及其对土壤真菌群落的影响,作为其环境安全性的评判依据。在此基础上,探索了利用工程菌中靶基因的引物在实时定量荧光PCR中的特异性,通过制定标准曲线快速定量检测甘蓝叶片上和环境土壤样品中工程菌含量的可能性,为工程菌剂的环境安全性评价和管控提供技术参考。主要结果摘要如下：工程菌剂BbHV8对甘蓝的整季保护效果与甲维盐匹敌在夏季由水稻和玉米作物包围的两地块中分别于4月2日和12日移载的甘蓝,很快诱集菜粉蝶等鳞翅目害虫前来产卵,田间菜青虫分别于4月22日和5月5日达到0.94±0.06和0.93±0.05头/株甘蓝(mean±SEM)的虫口密度,大大超过0.5头/株的经济危害阈值,故作为试验1和试验2用菌用药处理的起始日。两试验均包括4个处理,即BbHV8和WT的悬乳剂(1010个孢子/mL)100倍稀释液机动弥雾(用菌量均为1013个孢子/公顷)、5%甲维盐水分散粒剂按标签用量机动弥雾(标准化防)及配制孢子悬乳剂的乳油100倍稀释液机动弥雾(空白对照),每处理4个小区(6×11m),随机区组排列。试验1首次弥雾后间隔10天重复用菌用药,每隔5天抽样调查各处理中菜青虫、小菜蛾及蚜虫密度,第25天对照中菜青虫虫口达16.1±1.0头/株,而BbHV8、WT及甲维盐处理中分别为4.3±0.3、8.2±0.8和1.3±0.2头/株；首次药后20天,在BbHV8、WT、甲维盐和对照中小菜蛾虫口分别0.40±0.08、0.45±0.06、0.75±0.06及1.50±0.15头/株,第35天分别为0.15±0.05、0.28±0.09、1.95±0.64及0.13±0.05头/株。试验1中桃蚜初始密度为12.7±1.1头/株,首次药后15天甲维盐处理区蚜虫数量上升,第35天68.6±3.2头/株,而同期BbHV8处理区仅为10.9±1.5头/株。试验2的虫情与试验1相近,但对照处理菜青虫密度于第10天高达27.7±2.0头/株,于是分别在首次药后第5天和第15天重复用菌用药,最终试验2各处理的控虫效果优于试验1。由于两试验中都是暴食性菜青虫主导危害甘蓝,对照叶片损毁十分严重,且被损毁植株上的菜青虫就近转移到可取食的植株,使田间抽样调查的虫口数难以真实反映甘蓝受害状况,故调查时根据每株甘蓝叶片损失率分为5级,作为各处理相对防效和综合保护效果的评价依据。结果表明,试验1中BbHV8、WT和甲维盐历次抽样调查的相对防效分别为65～75%、15～58%和42～85%,平均防效分别为70.5%(±1.2)、30.9%(±6.1)和69.7%(±6.9)。试验2期间上述三处理的相对防效波动范围分别为49～80%、41～70%和51～90%,平均防效分别为74.5%(±4.3)、48.4%(±4.2)和83.5%(±5.5)。两个试验中,工程菌剂BbHV8与甲维盐的总体防效相似或相近,都显著高于WT。后者对害虫有一定的控制作用,但当菜青虫暴发时,难以及时控制虫口数量而对甘蓝予以有效保护。在试验2期间补充进行了两种菌剂相同用菌量下对暴食性4龄菜青虫的田间生物测定,测得BbHV8对4龄菜青虫的LT50和LT95分别为1.6(1.2～1.9)和4.5(4.3～4.8)天,而WT的LT50和LT９5分别为5.7(5.4～6.0)和11.5(10.6～12.8)天。这一结果说明,BbHV8毒杀4龄菜青虫50%和95%比WT分别快4天和7天。工程菌剂BbHV8的生态安全性在田间试验1和2期间的历次抽样调查中,BbHV8和WT菌剂处理对田间蜘蛛种群均未见负面影响,BbHV8由于较好保护了甘蓝,对田间蜘蛛种群有所助增,而甲维盐处理区蜘蛛数量一直低于菌剂处理区。在试验1最后一次用菌后次日、30天内每隔10天、半年内每隔一个月采集用菌和对照处理小区的表层土壤(0-5cm),带回实验室分别选择性分离培养球孢白僵菌与腐生丝状真菌并统计每克土样中的菌落形成单位(CFU)数量。结果显示,WT和BbHV8小区喷菌后次日每克土壤(干重)中球孢白僵菌CFU含量分别达3.1(±0.4)×104和2.1(±0.4)×104个,10天后降为4.3(±0.2)×103和3.2(±0.2)×103个,20天后降为3.8(±0.7)×102和4.1(±0.8)×102个,此后低水平徘徊,第120天后分别降至145±59和36±36个,再往后采集的土样中再未分离到球孢白僵菌。对照小区中仅在用菌后第20和30天的每克土样中分离到69±40和34±34个CFU,显示田间雨水等因素可导致白僵菌在地块内的轻微扩散。各批次土样中分离培养的腐生真菌CFU数量虽有季节性变化,但同批次样品中的CFU含量在两菌剂与对照处理间均无显著差异。将用菌后30天、60天和90天的球孢白僵菌土壤分离物各选24株共72株,每次来自BbHV8和WT小区的各12株,通过PCR检测它们是否携带杀虫蛋白基因vip3Aal基因。结果显示,来自用菌后30天WT小区的12株菌有2株呈阳性,而源于BbHV8小区的12株菌中也有2株呈阴性；用菌后60天的分离物鉴定结果,来自WT处理小区的白僵菌全为阴性,来自BbHV8小区的仅有1株呈阴性；用菌后90天的土样分离株中,来自BbHV8小区的全为阳性,而来自WT小区的全为阴性。显然,所用菌剂由于雨水泾流等原因在地块内存在扩散或相互串染的现象,但串染率不超过17%,且随着田间存留量的减少,串染比率降低,最后消失。用菌后90天BbHV8和WT小区的土壤分离株rHV8和rWT在SDAY平板上培养8天的产孢量分别比BbHV8和WT下降26.5%和32.4%,而且rHV8培养8天的菌落直径比BbHV8小32.2%；rWT虽比WT生长快9.1%,但菌落产孢量大幅下降。环境胁迫试验表明,BbHV8分生孢子耐UV-B辐射和45℃高温的能力分别比WT低19.2%和12.5%,rHV8比BbHV8分别低32.5%和38.9%,rWT比WT分别低39.7%和31.0%。总体看,BbHV8和Bb2860在释放环境中的生态适应能力呈下降趋势,且前者的生态适应能力弱于后者。工程菌剂BbHV8在释放环境中存留量的快速定量检测技术探索通过实时荧光定量PCR扩增融解曲线对靶基因vip3Aal引物特异性的分析建立扩增标准曲线,利用靶基因在BbHV8细胞中单拷贝的特性,以含分生孢子的甘蓝叶片和土壤样品的总DNA提取物为模板,探索建立甘蓝和土壤样品中孢子含量的快速检测方法。结果显示,菜叶上孢子的检测精度可达2个孢子/mm2,土壤中也可检出1-2个孢子/mg干土。然而,土壤成份复杂,影响目标DNA的提取效率和PCR扩增效率,样品检测偏差大于甘蓝叶片。综上所述,BbHV8转基因工程菌剂对甘蓝主要害虫的防治效果与甲维盐相似或相近,对非靶标的田间蜘蛛和土壤真菌群落无负面影响,其在释放环境中的存留量快速衰减,生态安全风险很低。更多还原

【Abstract】 Notoriously slow kill action of fungal cell-based insecticides often depresses commercial interest and genetically engineered mycoinsecticides can accelerate the kill action but associated ecological risk is of public concern. Since little effort has been made to assess control efficacy and ecological safety of transgenic mycoinsecticides under real field conditions, this study sought to fill the gap with the data of two35-day field trials performed in Jinhua, Zhejiang Province. Aerial conidia of a transgenic Beauveria bassiana strain (BbHV8) over-expressing a Bacillus thuringiensis vegetative insecticidal protein (Vip3Aal) for acquisition of fast per os virulence to caterpillars and its parental wild-type strain Bb2860(WT) were formulated for repeated sprays against cabbage pest complex, which was dominated by the cabbage butterfly (CBF) Pieris rapae, followed by the diamond-back moth (DBM) Plutella xylostella and the green peach aphid (GPA) Myzus persicae. Their control efficacies were compared with that of emarnectin benzoate (EB), a chemical insecticide commercially recommended for cabbage pest control. Their field fitness and effects on field spider community and native soil fungi were also evaluated during the cabbage growing season and several post-trial months respectively. An effort was made to establish a technical system that is potential for fast quantification of BbHV8conidia in cabbage foliage and soil samples by means of single-copy specificity of the target gene primers in quantitative real-time PCR (qRT-PCR). The results are summarized below.High field efficacy of BbHV8formulation competed with the EB efficacy in full-seson cabbage pest control. The two fields surrounded by rice and maize crops were transplanted with cabbage seedlings on April2and12respectively, attacking lepidopteran insects, such as CBF and DBM, for egg deposition. The mean CBF density (±SEM) reached0.94(±0.06) and0.93(±0.05) larvae per plant in the two fields on April22and May5, exceeding an economic injury level of0.5larvae per plant for initiatiing Trials1and2respectively. Each field trial included four treatments of four plots (6x11m), i.e.,100-fold aqueous dilution of emulasifiable BbHV8or WT formulation (1010conidia/ml) sprayed at the rate of1013conidia/ha, aqueous EB dilution sprayed at the labeled rate, and100-fold aqueous dilution of emulsion vector (for fungal formulation) sprayed as blank control (BC). All treatments and plots were arranged in terms of a randomized block design and monitored every5days for pest and spider densities until cabbage heads were ready for harvest. Repeated spray was carried out in Trial1at10-day internval, resulting in the respective CBF densities of16.1±1.01,4.310.3,8.2±0.8and1.3±0.2larvae per plant in the BC, BbHV8, WT and EB treatments on day25. The four treatments were infested with the DBM densities of1.50±0.15,0.40±0.08,0.45±0.06and0.75±0.06larvae per plant on day20and of0.13±0.05,0.15±0.05,0.28±0.09and1.95±0.64larvae per plant on day35respectively. GPA density in Trial1began from12.7±1.1aphids per plant and was consistently suppressed to a very low level by BbHV8throughout the tiral but increased to68.6±3.2aphids per plant in EB plots at the end of the trial. Cabbage plants in Trial2suffered from similar damages by main pest populations as seen in Trial1but the CBF density in BC was up to27.7±2.0larvae per plant on day10. For this reason, repeated sprays in Trial2were scheduled on days5,15and35respectively, resulting in somewhat better best control than that in Trial1.Since CBF overwhelmed foliage damage in both field trials and ageing larvae could transfer to adjacent plants from severely damage plants, pest population densities estimated by sampling could not reflect the true damage levels in the fields. Thus, damage of each sampled plant was scaled to five grades in terms of percent foliage loss due to ingestion by CBF larvae and the damge grades were used to compute control efficacy (%) relative to BC. As a result, the control efficacies of BbHV8, WT and EB fell in the respective ranges of65～75%,15～58%and42～85%during Trial1and were averaged as70.5%(±1.2),30.9%(±6.1) and69.7%(±6.9). The efficacies of the three treatments in Trial2spanned from49%to80%,41%to70%and51%to90%, resulting in the overall means of74.5%(±4.3),48.4%(±4.2) and83.5%(±5.5) respectively. In both trials, BbHV8provided a full-season protection from the damage of cabbage pest complex with an overall efficacy similar or close to that achieved by EB application. In contrast, WT showed some degree of control efficacy against the pest complex but the control was insufficient to protect the cabbage from the damage by large CBF population.A supplementary experiment was conducted to bioassay the virulence of BbHV8and WT to fourth-instar CBF larvae sprayed at the same rate as the two field trials during Trial2. Under the spray, LT50, LT95and associated95%confidence intervals were estimated as1.6(1.2～1.9) and4.5(4.3～4.8) days for BbHV8and5.7(5.4～6.0) and11.5(10.6～12.8) days for WT. These estimates indicate that BbHV8killed50%and95%of the tested larvae4-and7-day faster than WT. Ecological safety of transgenic BbHV8released in the field. During Trials1and2, spiders as nontarget predators of pests appeared on sampled plants were counted in situ and the counts on all sampling occasions revealed no negative effect of both BbHV8and WT on the spider population, which was even augmented by the BbHV8treatment but reduced by the chemical control.Surface soil samples were collected back to laboratory from the plots of BbHV8, WT and BC the day after the last spray of Trial1, every10day within the first month post-spray and every month from then on. Colony forming Units (CFUs) of B. bassiana and saprophytic fungi were isolated from the soil samples of each batch using two selective media respectively. As a result, CFU counts per gram of soil sample (dry weight) reached3.1(±0.4)×104and2.1(±0.4)×104in the soi samples taken respectively from WT and BbHV8plots the day after the last spray and declined to4.3(±0.2)×103and3.2(±0.2)×103on day10and to3.8(±0.7)×102and4.1(±0.8)×102on day20. From then on, CFU counts fluctuated at low levels and decreased to only145±59and36±36on day120but became undetectable afterwards. Interestingly, the counts69±40and34±34CFUs/g were found in the BC soil samples taken20and30days after the last spray respectively. Moreover, CFU counts of saprophytic fungi from different batches of soil samples showed seasonal fluctuation but did not differ significantly from one treatment to another within each batch.Seventy-two B. bassiana isolates recovered from the soil samples collected in the BbHV8and WT plots30,60and90days after the last spray were detected for the presence or absence of the target gene vip3Aal in their DNA samples. The target gene was present in33of the36isolates recovered from BbHV8plots but only in two of the36isolates recovered from Bb2860. Interestingly, the occasionally crossing presence of the target gene occurred only in the earlier soil samples. Since B. bassiana also occasionally recovered from the BC plots, our data indicated that the cabbage field lacked native B. bassiana strains and that the fungal sprays did cause a low frequency of crossing between treatments perhaps due to frequent light rains during the trial.Moreover, conidial yields of rHV8and rWT, two isolates recovered from the soil samples of BbHV8and WT treatments90days after the last spray, were26.5%and32.4%lower than BbHV8and WT, respectively, after8-day incubation on SDAY plates while rHV8grew32.2%slower than BbHV8. In stress assays, BbHV8conidia were19.2%and12.5%less tolerant to UV-B irradiation and wet-heat at45℃than the WT conidia while rHV8and rWT showed the reductions of38.9%and39.7%in conidial UV-B resistance and of32.5%and31.0%in conidial thermotolerance compared to BbHV8and WT respectively. Apparently, post-release ecological fitness was low for both BbHV8and WT and even lower for BbHV8.Rapid assessment of transgenic BbHV8conidia in cabbage foliage and soil samples by quantitative real-time PCR (qRT-PCR). High specificity of paired primers Vip3Aal-F/R to the target gene vip3Aal was determined with a melt curve in qRT-PCR, followed by establishing a standard curve for the amplication of vip3Aal from total DNA samples of10-105BbHV8conidia/ml. Since the target gene is signle-copied in each each transgenic cell, the number of transgenic conidia in each sample unit would equal to the number of DNA copies in the total DNA of each sumple unit. Thus, the established standard curve was used to assess conidial denisity in foliage and soil samples to which BbHV8conidia were quantitatively added. As a result, an accuracy of qRT-PCR assessment for the conidial density was2conidia/mm cabbage leave or1～2conidia/mg soil (dry weight). However, percent deviations between assessed and quantified conidial densities in soil samples were larger than those in foliage samples perhaps due to the complexity of soil components, which might affect efficiencies of both target DNA extraction and qRT-PCR amplification.Conlusively, the transgenic mycoinsecticide BbHV8not only competed with the chemical insecticide recommended for the control of cobbage insect pests dominated by P. rapae in two independent field trials but also had no adverse effect on nontarget spider population and native soil fungi. Taken together with a rapid decline rate of its residue in the soil of released field, BbHV8could be of little risk for environmental safety if it was commercially released.更多还原