What is the potential for artificial selection of insect biological control traits? A systematic review of their genetic variation, heritability, and evolvability

There is growing interest in trait selection for the improvement of insect biological control. Constraints from the Nagoya Protocol for sourcing novel biocontrol agents (BCAs) make optimization of accessible populations a higher priority. However, the effectiveness of selection programs remains unclear. This requires knowing whether biocontrol traits are heritable and whether captive populations have sufficient genetic variation for selection. To address this knowledge gap, we systematically reviewed reported values of broad-sense heritability (H²), narrow-sense heritability (h²), and evolvability (CV_a) for insect biocontrol traits, providing an overview of the potential for their genetic improvement while identifying the irregularities in the reporting of heritability data across studies. Our algorithmic search string focused on studies working with arthropod BCAs widely used in augmentative biocontrol and/or permitted for use in the EPPO region (European and Mediterranean Plant Protection Organization). Five thousand two hundred and nine initial search hits from the CAB Abstracts (n = 2321) and BIOSIS (n = 3442) databases recovered 277 relevant papers after screening, 94 of which included estimation methods and reported biocontrol traits as being heritable. However, over half (52) of these papers did not report any H², h², or CV_a values. There was strong indication of genetic variation (e.g., h² > 0.2) for numerous traits. Most data focused on pest suppression ability and mass reproduction, while little was reported on abiotic/biotic adaptation or ecological risk, despite their importance for biocontrol success. Contrary to expectations, heritabilities for complex life-history traits were equally high as those for morphology, whereas the heritability of insecticide resistance was relatively low. Despite relatively few data points and a large value range, this review demonstrates good potential for artificial selection of biocontrol traits and provides a single resource for collated data. We make recommendations for more consistent and expansive collection and repository of biocontrol trait data and to revise the Nagoya Protocol to facilitate breeding programs.