Do biological control agents adapt to local pest genotypes? A multi-year test across geographic scales

Parasite local adaptation has been a major focus of (co)evolutionary research on host-parasite interactions. Studies of wild host-parasite systems frequently find that parasites paired with local, sympatric host genotypes perform better than parasites paired with allopatric host genotypes. In contrast, there are few such tests in biological control systems to establish whether biological control parasites commonly perform better on sympatric pest genotypes. This knowledge gap prevents the optimal design of biological control programs: strong local adaptation could argue for the use of sympatric parasites to achieve consistent pest control. To address this gap, we tested for local adaptation of the biological control bacterium Pasteuria penetrans to the root-knot nematode Meloidogyne arenaria, a global threat to a wide range of crops. We measured the probability and intensity of P. penetrans infection on sympatric and allopatric M. arenaria over the course of four years. Our design accounted for variation in adaptation across scales by conducting tests within and across fields, and we isolated the signature of parasite adaptation by comparing parasites collected over the course of the growing season. Our results are largely inconsistent with local adaptation of P. penetrans to M. arenaria: in three of four years, parasites performed similarly well in sympatric and allopatric combinations. In one year, however, infection probability was 28% higher for parasites paired with hosts from their sympatric plot, relative to parasites paired with hosts from other plots within the same field. These mixed results argue for population genetic data to characterize the scale of gene flow and genetic divergence in this system. Overall, our findings do not provide strong support for using P. penetrans from local fields to enhance biological control of Meloidogyne.

Methods
To test for local adaptation of P. penetrans, we collected paired samples of M. arenaria and P. penetrans from six sites each year. Because the strength of local adaptation can vary with spatial scale, we collected samples from six plots within a single field in 2019 and 2020 and from six fields in 2021 and 2022. Each year, we compared the performance of P. penetrans when paired with sympatric and allopatric M. arenaria by measuring proxies for infection probability (attachment rate) and intensity (attachment load). In 2019 and 2020, we isolated the signature of parasite adaptation by comparing local adaptation of early-, mid-, and late-season parasites.

寄生虫本地适应（local adaptation）一直是宿主-寄生虫相互作用协同进化研究的核心议题之一。针对野生宿主-寄生虫系统的相关研究常发现，与异地（allopatric）宿主基因型配对的寄生虫相比，与本地同域（sympatric）宿主基因型配对的寄生虫表现更优。与之形成鲜明对比的是，生物防治系统中极少有实验能够验证生物防治寄生虫是否普遍在同域害虫基因型上表现更佳。这一知识缺口阻碍了生物防治项目的优化设计：若存在显著的本地适应，便可通过使用同域寄生虫来实现稳定的害虫防治。

为填补这一空白，我们针对生防细菌穿刺巴斯德芽菌（Pasteuria penetrans）对花生根结线虫（Meloidogyne arenaria）的本地适应能力开展了测试——后者是威胁多种作物的全球性有害生物。我们在四年间测定了穿刺巴斯德芽菌在同域与异地花生根结线虫上的感染概率与感染强度。本实验设计通过在田块内部及跨田块开展测试，考量了不同空间尺度下的适应能力差异，并通过比较生长季不同时段采集的寄生虫特征，分离出寄生虫本地适应的信号。

我们的研究结果整体上不支持穿刺巴斯德芽菌对花生根结线虫的本地适应：在四年中的三年里，寄生虫在同域与异地组合中的表现并无显著差异。然而在其中一年，与来自同域田块的宿主配对的寄生虫，其感染概率较与同一块田内其他样地宿主配对的寄生虫高出28%。这些混合结果表明，亟需通过种群遗传学数据来表征该系统中的基因流规模与遗传分化程度。总体而言，本研究并未为使用本地田块的穿刺巴斯德芽菌来强化花生根结线虫的生物防治提供有力支撑。

方法
为测试穿刺巴斯德芽菌的本地适应能力，我们每年从6个采样点采集花生根结线虫与穿刺巴斯德芽菌的配对样本。由于本地适应的强度可能随空间尺度变化，我们在2019年与2020年从同一块田的6个样地（plot）中采集样本，2021年与2022年则从6个独立田块采集样本。每年，我们通过测定感染概率的替代指标附着率（attachment rate）与感染强度的替代指标附着负荷（attachment load），比较穿刺巴斯德芽菌与同域、异地花生根结线虫配对时的表现。在2019年与2020年，我们通过比较生长季早期、中期与晚期采集的寄生虫的本地适应特征，进一步分离出寄生虫本地适应的信号。