Alan Raybould, Ecological Sciences, Syngenta Risk assessments for weediness of genetically modified crops 遗传改良作物杂草化的风险评估 1. 1. Agricultural habitats 2. Non-agricultural habitats 1.农业环境 2. 非农业环境 Alan Raybould, Ecological Sciences, Syngenta

Typical management objective from the Plant Protection Act (USA) Weediness in agricultural habitats – problem formulation 农业环境中的杂草化 – 问题定义 Typical management objective from the Plant Protection Act (USA) “safeguard agriculture and natural resources from the risks associated with the entry, establishment, or spread of animal and plant pests and noxious weeds to ensure an abundant, high-quality, and varied food supply” 美国植物保护条例的典型管理目的 “保护农业和天然资源,减少动物,害虫和有害种子传播的风险,确保充分的,高质量的,丰富的食物供应.”

Weediness in agricultural habitats – problem formulation 农业环境中的杂草化 – 问题定义 Assessment endpoints may be derived from these objectives Abundance – crop yield High-quality – nutritional value of crops Varied – diversity of crops 评估终点来源于下列目标 大量 – 作物产量 高质量 – 作物营养含量 变化性 – 多样作物种类

Weediness in agricultural habitats – conceptual model 农业环境中的杂草化– 概念模型 How may a GM crop affect these endpoints? J. Orson 1993. The penalties of volunteer crops. Aspects of Applied Biology 35: 1-8 Review of harm associated with volunteer crops in UK Three categories of harm Competition Green Bridges (hosts of disease when no crop present) Contamination GM作物怎样影响这些终点? J. Orson 1993. The penalties of volunteer crops. Aspects of Applied Biology 35: 1-8 英国自生作物危害性回顾 三种危害性 竞争 绿桥 (无作物存在时疾病的宿主) 污染

Competition 竞争 Slow growing crop follows fast growing crop 慢速生长作物跟随快速生长作物 Onions sown after potatoes 洋葱在土豆之后播种 Yield no weed control: 0-13 tonnes/hectare Yield hand weeding: 20-62 tonnes/hectare Best herbicide control: 53 tonnes/hectare 无杂草控制的产量: 0-13 吨/公顷 有杂草控制的产量: 20-62 吨/公顷 最好的除草剂控制: 53 吨/公顷

Green bridges 绿桥 Yellow rust infection of wheat 小麦的黄锈病感染 Yarham DJ, Gladders P. 1993. Effects of volunteer plants on crop diseases. Aspects of Applied Biology 35: 75-82 中文表请见下页

Green bridges 绿桥

Contamination 污染 Loss of quality if seed (or other parts) of volunteer harvested with crop Different crops e.g. peas after linseed Linseed pods similar size to peas – difficult to separate Factories rejected peas sown in rotation with linseed Same crop High glucosinolate/low glucosinolate oilseed rape Wheat for animal feed/wheat for milling 如果自生植物种子(或其他部分)和作物一起被收割将影响作物的质量 不同作物 例如 豌豆 亚麻子 亚麻子的豆荚与豌豆大小相似很难分离 厂商一般不接受在亚麻子之后播种得到的豌豆 相同作物 高芥甙/低芥甙油菜籽 磨粉用/饲料用小麦

Weediness in agricultural 农业环境中的杂草化 habitats – conceptual model – 概念模型 Crops harm assessment endpoints of yield and nutritional value Yield – competition and green bridges Nutrition – contamination GM crops can harm these endpoints in the same way Any other ways in which GM crops can harm these endpoints? “ 作物危害性评估的产量和营养价值终点 产量 – 竞争和绿桥 营养 – 污染 GM作物可以以相同的方式来危害这些评估终点 GM作物还有其他危害方式么?

Weediness in agricultural 农业环境中的杂草化 habitats – conceptual model – 概念模型 The transgenic process presents no new categories of risk compared to conventional methods of crop improvement” National Research Council. 2002. Environmental Effects of Transgenic Plants: the Scope and Adequacy of Regulation Competition, green bridges and contamination are sufficient “转基因过程和传统的作物改良方式相比并没有引进新的风险” National Research Council. 2002. Environmental Effects of Transgenic Plants: the Scope and Adequacy of Regulation 竞争,绿桥和污染已经足够了

Weediness in agricultural 农业环境中的杂草化 habitats – conceptual model – 概念模型 Current agricultural practice is deemed to pose acceptable risk to the assessment endpoints Risk of weediness of GM crops can therefore be assessed by evaluating whether they show increased weediness: More competitive than the non-GM crop Greater probability of being a green bridge than the non-GM crop More frequent or more serious contamination than non-GM crop 现行的农业实践被认为会引起一些可接受的评估终点风险 因此,GM作物的杂草化风险可以用以下几点来评估: 比非GM作物更有竞争性 比非GM作物更有可能成为绿桥 比非GM作物的污染持续性更强或更严重

Weediness in agricultural 农业环境中的杂草化 habitats – conceptual model – 概念模型 The null hypothesis is that the GM crop is no more weedy than the non-GM crop: this is the risk hypothesis A secondary risk hypothesis is that GM crop x wild plant hybrids are no more weedy than non-GM hybrids in agricultural habitats 初始假设是GM作物不会比非GM作物引进更多的杂草:这是风险假设 次级假设是GM作物与野生种杂交后代在农业环境中不会比非GM作物引进更多的杂草

Testing the risk hypothesis 检验风险假设 -competition – 竞争 Traits that may indicate increased competition Appendix II Canada and United States “Bilateral on Biotechnology” www.inspection.gc.ca/english/plaveg/bio/usda/appenannex2e.shtml Reproduction and survival characters Life span, vegetative biomass, over-wintering capacity, flowering behaviour, seed production, seed dormancy, germination, seedling survival, outcrossing frequency, pollen viability, dispersal ability 一些特征表明竞争增加 附录II 加拿大和美国 “双边生物技术” www.inspection.gc.ca/english/ plaveg/bio/usda/appenannex2e.shtml 繁殖和生存特征 生存期限, 植物数量, 越冬能力, 开花行为, 种子产生, 种子休眠, 萌发, 幼苗存活, 异型杂交频率, 花粉发育能力, 散播能力

Testing the risk hypothesis 检验风险假设 -competition – 竞争 Adaptation to stress Biotic: pathogens,herbivores, other plants Abiotic: atmospheric, pollutants, nutrient, deficiency, temperature extremes, drought, flood Pesticides 压力适应 生物压力: 病原, 草食动物, 其他植物 非生物压力: 大气污染, 营养不足, 温度极限, 干旱, 洪水 杀虫剂

Testing the risk hypothesis 检验风险假设 -competition – 竞争 Reproduction and survival characters are measured in agronomic field trials Comparison of GM crop and non-GM, near-isogenic lines Several genetic backgrounds Multiple locations – represent range of proposed cultivation Two years Data on efficacy and performance are the main objectives Data are suitable for safety 遗传和生存特征可以在农业试验田中测得 GM和非GM作物比较, 相似-相同基因列 几个遗传背景 不同位置 – 代表不同耕作区域 两年 功能和表现上的数据是主要的目标 数据对安全性来说是合适的

Testing the risk hypothesis 检验风险假设 -competition – 竞争 Biotic stress-tolerance parameters are also measured Pests and disease Weed infestation 生物压力-耐受参数也要测试 害虫和疾病 杂草丛生

Testing the risk hypothesis 检验风险假设 Will the intended effect of the transgene increase weediness? Are volunteer weeds controlled by the pest protected against? Agricultural equivalent of ecological release Usually no special studies are carried out to assess this 转基因的作用会增加杂草化么? 自生杂草会因为害虫控制而被保护起来么? 生态释放的农业平衡 通常没有特殊的研究来执行这个评估

Testing the risk hypothesis 检验风险假设 Field trials are monitored for unusual frequency of volunteers Also arguments from timing and effects of pest attack If pests attack mature plants, unlikely to affect volunteering Increase in yield is unlikely to lead to more volunteers if seeds have no dormancy and plants cannot over-winter 在试验田监测异常的自生植物生长频率 一些害虫攻击的时间和效果的争论 如果害虫攻击成熟植株, 那么就不会攻击自生植物 如果种子没有休眠而且植物没有越冬,增加作物产量并不会导致自生植物的增加

Testing the risk hypothesis 检验风险假设 -competition – 竞争 If exposure assessment indicates crop x weed hybrids are possible Could test the competitive ability of GM and non-GM hybrids A lot of work to generate hybrids Need to conduct a separate trial from the agronomic trial 如果暴露评估显示作物与杂草杂交是可能的 那么就能够测定GM作物与非GM作物杂交后代的竞争能力 产生杂交后代需要大量工作 需要一个与农用田分离地试验田

Testing the risk hypothesis 检验风险假设 -competition – 竞争 Regard the hybrids as an additional genetic background Use agronomic data to test for transgene x genotype interactions If no interaction, crop data predict hybrid behaviour 把杂交后代作为另外的遗传背景 利用农业数据测试转入基因和原基因相互作用 如果没有相互作用, 作物数据可预测杂交行为

Testing the risk hypothesis 检验风险假设 -competition – 竞争 Measurements of abiotic stress tolerance are not usually made Indirectly assessed by performance in multiple locations Sometimes specific studies are carried out 并不需要经常监测非生物压力的耐受 间接评估不同区域的表现 有时候需要特殊的研究

Testing the risk hypothesis 检验风险假设 -competition – 竞争 E.g., maize over-wintering study for Japan GM and non-GM seeds are germinated Seedlings are placed in a growth chamber at ca. 0oC Survival is measured Simulates conditions required for over-wintering 例如, 在日本的玉米越冬研究 GM作物和非GM作物种子萌发 幼苗置于0oC 培养箱 测试存活量 模拟越冬条件

Testing the risk hypothesis 检验风险假设 -competition – 竞争 测试除草剂耐受作物的功效 和非GM作物相比显示出很高的耐受能力 在下列作物中杂草化风险较高 Herbicide tolerant crops are tested for efficacy Obviously show higher tolerance than the non-GM crop Risk of greater weediness in following crop

Testing the risk hypothesis 检验风险假设 -competition – 竞争 E.g., glyphosate tolerant maize and soybean rotations Use risk management to reduce risk Cultivation to remove volunteers Rotate herbicides, use tank mixes These practices will also reduce risk of the evolution of tolerance in weeds (by selection or gene flow) 例如, 草苷膦耐受玉米和大豆 风险管理的应用可降低风险 适当耕作以清除自生植物 轮换除草剂,使用罐子混合 这些方法可以降低杂草耐受性进化的风险

Testing the risk hypothesis 检验风险假设 -competition -绿桥 Null hypothesis is unchanged probability of being a green bridge No change in persistence and abundance after harvest No change in susceptibility to pests and disease (other than any intended increase in resistance due to the transgene) 初始假设:形成绿桥的可能性是不变的 在收割后在持续性和丰富性上没有变化 对害虫和疾病的敏感性上没有变化 (与转基因产生的抗性增加不同)

Testing the risk hypothesis 检验风险假设 -competition -绿桥 Estimates of these parameters are made in the agronomic studies Therefore data requirements are the same as for competition 对这些参数的评估建立在农业研究的基础上 因此必须的数据和竞争性研究是相同的

Testing the risk hypothesis 检验风险假设 -contamination -污染 There are two sources of increased contamination Increased abundance of the contaminating crop Increased hazard of the contaminant Increased abundance is assessed from the agronomic data Increased hazard is assessed using a variety of studies Composition study Developmental study Toxicity studies on the GM proteins 有两种途径可能增加污染程度 被污染作物的丰富性增加 污染的危害性增加 农业数据可以评估丰富性的增加多种研究可以评估危害性的增加 成分分析 发展研究 GM蛋白的毒性研究

Testing the risk hypothesis 检验风险假设 -contamination -污染 Composition study Compares composition of GM and non-GM crops Numerous parameters are compared Nutrients, and anti-nutrients and toxins where relevant 成分分析 比较GM和非GM作物的成分 需要比较许多参数 相关的营养, 非营养和毒性物质

Testing the risk hypothesis 检验风险假设 -contamination -污染 OECD has published guidance on which parameters to measure Crops include oilseed rape, soybean, maize, sugar beet, potatoes, wheat, rice, cotton, barley and alfalfa Index of documents is on the OECD website: www.oecd.org/document/9/0,2340,en_2649_34391_1812041_1_1_1_1,00.html OECD出版了参数测量指导 包括油菜,大豆,玉米,甜菜,土豆,小麦,水稻,棉花,大麦和苜蓿等作物 文件索引在OECD的网页上: www.oecd.org/document/9/0,2340,en_2649_34391_1812041_1_1_1_1,00.html

Testing the risk hypothesis 检验风险假设 -contamination -污染 Composition also assessed indirectly in long-term feeding studies Animals fed material of GM plant Compared with animals fed non-GM material 42-day broiler chicken study Growth and development 90-day rat rat study Biochemical, physiological and histological measures 成分评估也可以在长期饲养研究中间接进行 GM植物作为动物饲料 与非GM植物作为饲料的动物相比 42天布罗勒鸡研究 生长与发育 90天大鼠研究 生化,生理和组织学方法测试

Testing the risk hypothesis 检验风险假设 -contamination -污染 Developmental study Determines where, when and at what concentration the GM proteins are expressed in the GM crop Allows prediction of the concentrations of the GM proteins that may occur under different contamination scenarios Environmental fate assessment E.g., if proteins are confined to grain, contamination will not occur if volunteers are removed before flowering 发育研究 确定GM蛋白在作物中表达的时间位置和浓度 可以预测发生在不同污染模式下的GM蛋白浓度 环境归趋评估 例如, 如果蛋白质局限于谷物,那么如果在开花前清除自生植物的话,污染就不会发生

Testing the risk hypothesis 检验风险假设 -contamination -污染 Toxicity of the GM protein is tested for human safety assessment Acute rat or mouse studies at very high dose No observable effect level (NOEL) is determined Usually highest dose in the study 人类健康评估测试GM蛋白的毒性 极高浓度大鼠小鼠急性研究 可以测得非可见作用水平 (NOEL) 在研究中通常使用最高浓度

Testing the risk hypothesis 检验风险假设 -contamination -污染 If the protein is deemed safe in a food crop, it should be regarded as safe when the crop is a contaminant in a different crop In a non-food crop, compare the NOEL with the predicted levels in food crops should contamination occur Also consider exposure via hybrids with wild relatives if relevant Estimate of risk = NOEL/predicted level 如果在食用作物中蛋白被认为是安全的,那么当这种作物污染其他作物时,也是安全的 在非食用作物中,比较NOEL与可能发生污染的食用作物预测水平 还需考虑通过与野生物种杂交的后代的暴露性 评估的风险值 = NOEL/预测水平

Summary risk assessment for Summary risk assessment for 农业环境中杂草化 weediness in agricultural habitats 风险评估概述 Safety is demonstrated if the risk hypothesis that the GM crop is no more weedy than the non-GM crop is not falsified: Agronomic field trials comparing GM and non-GM crop No significant difference in reproduction and survival No transgene x genotype interaction (if hybrids are possible) No significant difference in disease and pest susceptibility (other than any intended increase in resistance due to the transgene) 安全研究证明了GM作物不会比非GM作物产生更多的杂草这一风险假设: 农业试验田比较GM作物和非GM作物 在繁殖和生存率上没有显著差异 没有转基因品种和原基因型品种相互作用(如果有可能出现杂交后代的话) 在疾病和害虫敏感性上没有显著差异(由转基因引起的抗性没有明显增强)

Summary risk assessment for Summary risk assessment for 农业环境中杂草化 weediness in agricultural habitats 风险评估概述 Abiotic stress-tolerance study comparing GM and non-GM crop No significant difference in survival 在非生物压力耐受性方面比较GM作物和非GM作物 在存活率上没有显著差异

Summary risk assessment for Summary risk assessment for 农业环境中杂草化 weediness in agricultural habitats 风险评估概述 Composition study comparing GM and non-GM crop No significant difference in composition Feeding studies comparing GM and non-GM crop No significant difference in growth, survival etc. 在成分分析方面比较GM作物和非GM作物 在成分上没有显著差异 在饲料研究方面比较GM作物和非GM作物 在生长和存活率等方面没有显著差异

Summary risk assessment for Summary risk assessment for 农业环境中杂草化 weediness in agricultural habitats 风险评估概述 Toxicity study and environmental fate assessment NOEL > predicted worst-case exposure via contaminated crop If these results are obtained, weediness risk is minimal 毒性研究和环境归趋评估 NOEL >预测的通过污染作物暴露的最坏结果 如果得到了这些结果,杂草化风险将达到最小化,杂草化风险是最小的

Risk management of weediness 杂草化的风险管理 Herbicide tolerance: HT crops may pose increased weediness risk Use agronomic practices to minimise risk Rotate herbicides, mechanically remove volunteers etc. 除草剂耐受: HT作物可能会引起杂草化风险的增加 应用农业实践来降低风险 使用不同的除草剂,用机械清除自生植物

Risk management of 杂草化的风险管理 Weediness Pharmaceutical proteins: If NOEL > predicted exposure via contaminated crop, there may be an economic risk, even though no safety risk Rigorous control of volunteers 药用蛋白: 如果NOEL > 预期的通过污染作物的暴露量, 那么即使没有安全风险,也会有经济风险 对自生植物严格控制

Weediness in non-agricultural. 在非农业环境中的 habitats – problem formulation Assessment endpoints for weediness in non-agricultural habitats Pimentel (2001) reviewed harm from invasive alien plants Displacement of native plants species Physical changes: reduced water supply, increased frequency of fires and changed nutrient cycles 在非农业环境中的杂草化评估终点 Pimentel (2001) 评论外来入侵植物的危害性 取代了本地植物 物理上的改变:减少了水供给,增加了火灾发生的可能性,改变了营养循环

Weediness in non-agricultural. 在非农业环境中的 habitats – problem formulation Detrimental effects on recreation: a particular problem with aquatic plants that affect fishing, boating, swimming etc. Loss of yield in semi-natural pastures Costs of control Use these assessment endpoints to assess risks of GM crops 对人们休闲娱乐的影响: 特别的水生植物问题会影响人们钓鱼,划船,游泳等娱乐 半自然牧场产量减少 成本控制 应用这些评估终点来评估GM作物的风险性

Weediness in non-agricultural. 在非农业环境中的 habitats – problem formulation How may a GM crop affect these endpoints? By being an invasive plant itself By hybridising with a wild relative, producing an invasive plant GM作物是怎样影响这些终点的? 自身作为入侵植物 与野生植物杂交,产生入侵物种

Weediness in non-agricultural. 在非农业环境中的 habitats – problem formulation Current agricultural practice is deemed to pose acceptable risk to the assessment endpoints Crops are rarely invasive of natural habitats Crop traits are disadvantageous in natural habitats No dormancy, low dispersal, highly palatable seeds etc. 现在的农业实践被认为会产生适合的风险评估终点 作物不会入侵自然环境 作物的遗传特征不利于在自然环境中生存 没有休眠期,低传播率,种子非常适合食用等特征

Invasive plants are horticultural not agricultural 入侵植物是园艺中的,并不是农业上的 喜马拉雅香胶树 日本书草 Japanese knotweed Himalayan balsam 北美杜鹃 巨豚草 Rhododendron Giant hogweed

Weediness in non-agricultural. 在非农业环境中的 habitats – problem formulation Current agricultural practice is deemed to pose acceptable risk to the assessment endpoints The null hypothesis is that GM crops and their hybrids with wild relatives are no more invasive than the non-GM equivalents Test the risk hypothesis of no greater weediness 现在的农业实践被认为会产生适合的风险评估终点 初始假设GM作物及其与野生亲缘杂交后代并不会比非GM作物有更强的侵略性 测试少量杂草的风险假设

Testing the risk hypothesis 检验风险假设 The agronomic study is useful for predicting weediness in non-agricultural habitats GM crop and hybrids are unlikely to be more invasive than the non-GM crop if the following applies No significant difference in reproduction and survival No transgene x genotype interaction No significant difference in disease and pest susceptibility 农艺研究对于预测在非农业环境中的杂草化是有效的 如果出现以下情况表明GM作物及其杂交后代不会比非GM作物表现出更强的入侵性 在繁殖和生存率上没有显著差异 没有转基因和原基因相互影响 在疾病和害虫敏感性上没有显著差异

Testing the risk hypothesis 检验风险假设 To complete the risk assessment need to consider whether the intended effect of the transgene will lead to ecological release 为了完成风险评估,我们需要考虑是否转基因会导致生态释放

Predicting ecological release 预测生态释放 For crops that rarely form feral populations, ecological release can be assessed without specific studies Lack of invasiveness is likely to be due to restricted dispersal, lack of dormancy and inability to establish on undisturbed ground 对于很少产生野化种群的作物来说, 不用特殊的研究也可以测得生态释放 入侵的减少可能是由于种子传播受到限制,缺乏休眠期,和没有形成无干扰区的原因

Predicting ecological release 预测生态释放 Increases in pest and disease resistance are highly unlikely to change ability to disperse and persist If no increase in reproduction or survival traits in the agronomic trial, risk of ecological release is minimal 对害虫和疾病抗性的增强很可能会改变种子传播和持续的能力 如果在农业试验中繁殖和生存的特征没有增强,生态释放的风险是最小的

Predicting ecological release 预测生态释放 For crops that frequently form feral 对于经常会产生野化种群的作物来, populations, may need to assess 说需要用特殊的研究方法来评估生 ecological release with specific studies 态释放 Could simulate effect of transgene 可以模拟转基因的效果 E.g., test for ecological release of insect 例如,抗虫油菜生态释放测试 resistant oilseed rapeSow seed into habitat 将种子播种到种子可以自然散布的环境中 where seed may disperse naturally 保护种子不被害虫攻击(如施用杀虫剂) Protect some replicates from insect attack 比较新补充的植物和未受保护的植物 (e.g., insecticide) 如果没有差异, 生态释放风险很低 Compare recruitment of plants with that in unprotected replicates If no difference, risk of ecological release is low

Predicting ecological 预测杂交后代的 release in hybrids 生态释放 Identify species and locations where hybrids likely Exposure assessment described in previous talk Could carry out studies simulating the effect of the transgene May not be possible in natural populations Application of pesticides may be prohibited May not be possible to exclude pests or disease effectively 确定杂交后代的种类和可能出现的区域 在前面谈到了暴露评估 可以模拟转基因的作用来进行研究 可能对于自然种群来说不可能 杀虫剂的施用是被禁止的 不可能排除害虫或疾病的影响

Predicting ecological 预测杂交后代的 release in hybrids 生态释放 Alternative approaches are possible Identify phenomena that must happen for ecological release Test for these phenomena under controlled conditions Example using turnip mosaic virus resistance in oilseed rape 有其他可供选择的方法 确定生态释放必定发生的现象 在一定条件下测定这些现象 例如在油菜中使用芜菁花叶病毒抗性基因

Species studied 物种研究 油菜 野生甘蓝 野生芜菁 黑芥子 Brassica napus – oilseed rape B. oleracea – wild cabbage 野生芜菁 黑芥子 B. nigra – black mustard B. rapa – wild turnip

Tiered tests for hazard 危害性的分级测试 For “ecological release” to occur The plant must be susceptible to the virus The virus must reduce plant survival or reproduction The plant must become infected in the field Population dynamics must be density-independent Can design simple tests for existence of these phenomena 生态释放发生 植物必须对病毒敏感 病毒必须要减少植物的存活率或繁殖能力 植物在田地内必须被感染 种群动态必须是密度依赖型 可以为了检验这些现象的存在而设计简单的测试

Tiered tests for hazard 危害性的分级测试 Tier I: inoculate plant with virus in the lab If immune, no hazard, stop testing; not immune, tier II 阶段 I: 在实验室中将病毒输入植物体 如果有免疫, 没有危害性, 停止测试; 没有免疫, 进入阶段 II

Tiered tests for hazard 危害性的分级测试 Tier II: measure survival and reproduction of infected plants If unchanged, no hazard, stop testing; if lower, tier III 阶段 II: 测试被感染植物的存活率和繁殖能力 如果没有变化, 没有危害性, 停止测试; 如果降低,进入阶段III

Tiered tests for hazard 危害性的分级测试 Tier III: measure infection in the field If no infection, no hazard, stop testing; if infection, tier IV 阶段 III: 在田间监测感染 如果没有感染, 没有危害性,停止测试; 如果感染,进入阶段 IV

Tiered tests for hazard 危害性的分级测试 Tier IV: test for density independent population growth Various methods Sow seeds, exclude virus vectors etc 阶段 IV: 密度依赖型种群的生长测试 不同的方法 播种, 排除病毒载体

TuMV and wild Brassicas 在英国的TuMV和 in the UK 野生芸苔 Minimal risk of gene flow from TuMV-resistant oilseed rape B. nigra: negligible gene flow, no infection in field B. rapa: high gene flow, no infection in the field B. oleracea: low gene flow, no effect on population size TuMV抗性油菜基因流动的最小风险 B. nigra: 基因流动可忽略, 没有田间感染 B. rapa: 高基因流动,没有田间感染 B. oleracea: 低基因流动, 在种群范围没有影响

Summary risk assessment for Summary risk assessment for 在非农业环境中杂草化 weediness in non-agricultural habitats 风险评估概述 Safety demonstrated if hypothesis that GM crop and its hybrids are no more weedy than the non-GM equivalents is not falsified: Agronomic field trials comparing GM and non-GM crop Compare reproduction and survival Test for transgene x genotype interaction (if hybrids are possible) Compare disease and pest susceptibility (other than any intended increase in resistance due to the transgene) 安全研究证明了GM作物不会比非GM作物产生更多的杂草这一风险假设: 农业试验田比较GM作物和非GM作物 比较繁殖能力和存活率 测试转基因和原基因的相互作用 (如果有杂交后代) 比较对害虫和疾病的敏感性 (由转基因引起的抗性没有明显增强)

Summary risk assessment for Summary risk assessment for 在非农业环境中杂草化 weediness in non-agricultural habitats 风险评估概述 If no significant differences, risk of increased weediness in non-agricultural habitats is minimal Provided tests show minimal risk of ecological release 如果没有显著差异,在农业环境中杂草化增加的风险是最小的 测试显示了生态释放的最小风险

Summary risk assessment for Summary risk assessment for 在非农业环境中杂草化 weediness in non-agricultural habitats 风险评估概述 Ecological release assessment: Crops that rarely form feral populations No differences in agronomic data are sufficient to show low risk Crops that form feral populations May need extra data for stress tolerance traits E.g., sow seeds and simulate effect of transgene No difference between treatment and control shows low risk 生态释放评估: 不经常形成野化种群的作物 农业数据显示没有显著的表示低风险的差异 形成野化种群的作物 需要更多的压力耐受特征的数据 例如, 播种和模拟转基因效果 处理组和对照组之间没有显著差异来证明低风险

Summary risk assessment for Summary risk assessment for 在非农业环境中杂草化 weediness in non-agricultural habitats 风险评估概述 Hybrids Test for phenomena necessary for ecological release If phenomena are not detected, low risk is demonstrated 杂交后代 测试生态释放必要的现象 如果没有监测到这些现象,则证明风险性低