Final Report Summary - COLBICS (Intersectoral Collaborations to Boost Research and Development Dynamics in Biological Control of Agricultural pests) The project COLBICS (2013-2016; www.colbics.eu) sets a framework for long-term collaborations between three academic organisations and three companies, to generate innovations in biological control of agricultural pests. COLBICS academic participants are the “Institut National de la Recherche Agronomique” (INRA France), the Universitat Politècnica de Valencia (UPV, Spain) and the Pontificia Universidad Católica de Chile (UC, Chile). The three companies involved are Biobest Belgium NV (Biobest); Invivo Agrosolutions (IAS, France) and Anasac Chile SA (Anasac).COLBICS relies on the creation of three teams, each mixing researchers from one company and at least two academic partners. These teams carry out R&D activities along three axes. The first axis is the use of molecular biology to characterize crop pests and their natural enemies for applications such as the exploration of biodiversity to look for new biological control agents, the setup of quality control procedures, or the development of pest or natural enemy identification tools. The second axis is the improvement of production methods and products. COLBICS addresses three challenges along this axis: (i) the improvement of production methods with new protocols and diets; (ii) the optimisation of biocontrol agents through the exploitation of the available intraspecific diversity; and (iii) a consortium-wide evaluation of the actual impacts of genetic processes in biocontrol agent mass-reared populations. The third axis supports the integration of biological control agents in field situations. It is developed in COLBICS by two means: (i) field-evaluation of biocontrol products and release methods gathering data enabling to adjust the use of the products and to promote them with farm-advisors and producers; (ii) second, the setup of decision support tools facilitating the relevant use of biocontrol products.Molecular characterisation activities generated large amounts of data on biological material internal to the biocontrol companies or collected in field-surveys. The companies could obtain DNA barcodes for all their biological control agents. Uncertainties about the taxonomic identification of species have been clarified. For example, a ladybeetle commercialized in Chile under the name Coccidophilus citricola proved to be a misidentification of the native species Coccidophilus transandinus. In Europe, large-scale comparisons of field and commercial populations supposedly belonging to the widely used biocontrol agent Macrolophus pygmaeus were actually found to be a set of several highly divergent and partially reproductively isolated genetic clusters. Field sampling of Trichogramma wasps on Lepidoptera eggs and natural enemies of scale insects also provided insights into several species complexes of potential economic interest. In addition, the teams developed several molecular tools for the rapid detection and identification of pests or natural enemies. This was in particular the case for mealybug species that identification is difficult and that control often relies on specialist natural enemies. These activities also directly led to the development of three new products by the involved companies.In the second axis developed in COLBICS, new production methods were developed for a set of biocontrol agents (Coccinellidae, Chrysopidae, predatory mites) or hosts used for biocontrol agent production (Lepidoptera), drastically decreasing their cost of production. In five species, a consortium-wide study of the impacts of genetics (inbreeding, outbreeding) in mass-rearings of biocontrol agents was carried out. While little or no evidence of inbreeding depression was found for the species Chrysoperla comanche, Cryptolaemus montrouzieri, Allotropa burrelli and Ephestia kuhniella, evidence of sensitivity to inbreeding was observed in Macrolophus pygmaeus. Opportunities of genetic improvement via hybridization and selection were revealed in Macrolophus sp. and Trichogramma sp. However, risks associated to the practice of mixing field-collected populations, caused by the occurrence of unknown cryptic and reproductively isolated populations, have also been evidenced in Macrolophus sp. and Chrysoperla sp. The teams also worked on new approaches to study more efficiently the impact of genetic processes in industrial settings, taking profit from the last generation of genomic tools.Field-efficiency evaluations were carried out for several biocontrol products. Contrasted results were obtained, ranging from no detection of short-term impacts to significant decreases of pest populations. For example, in Spanish vineyards: no short-term impact of releases of Chrysoperla carnea and Anagyrus pseudococci could be detected on mealybug pests. On the other hand, in citrus orchards augmentative releases of the soil-dwelling mite Hypoaspis aculeifer resulted in lower fruit damage caused by the invasive thrips Pezothrips kellyanus but no effect of Stratiolaelaps scimitus releases was detected on this pest; the application of food complements such as NutrimiteTM (Biobest) increased the population density of the local beneficial phytoseiid mite populations; the use of new sugar dispensers developed by Biobest altered the mutualism between ants and mealybugs impactingant activity and decreasing the density of mealybugs; no short term effect of Aphytis melinus was detected on the scale insect Aonidiella aurantii. In Chile, experiments using Rhizobius lophantae revealed no short-term impact of releases on scale insects in avocado trees. In France and Italy, experiments using Trichogramma brassicae on maize revealed differences of field dispersal among different populations and strains. In terms of decision-guiding tools, the COLBICS team Anasac-UC-INRA has worked on the improvement of a Pest-Monitoring Service commercialized by the Anasac subsidiary Xilema SpA. R&D has led to new user-friendly interfaces for both data inputs (for operators in charge of the monitoring) and outputs under the form of web-interfaces exploiting monitoring data to generate alerts and recommendations to end-users.COLBICS has also been rich in training and networking activities. Most training efforts have focused on foreign languages to optimize communication, and on molecular characterisation that took a central place in COLBICS. Networking activities included key events (notably, two international workshops or conferences open to researchers from outside the consortium were organised in 2014 and 2016) and multiple meetings with partners at each level of the biocontrol value chain. Networking has been particularly fruitful and generated the emergence of more than 10 new national or international project proposals reinforcing or extending COLBICS activities, 5 of each have been financed and are ongoing. COLBICS had strong impacts on the career of the researchers involved, be it in terms of evolution of activities for researchers with permanent positions or training and employment for researchers without permanent position (for instance, the employment rate of the recruited experienced researchers involved in Colbics was 100%).Diffusion of results and information about COLBICS activities were carried out via the publication of articles in newspapers, specialist national journals and peer-reviewed scientific journals. Many meetings were also organized with stake-holders (from growers and farm advisors to representatives of ministries in charge of public policies). Finally, COLBICS researchers have familiarized the public with biological control, via outreach activities of different types, such as events in schools, with non-governmental organisations, stands large national exhibitions or participations to the creation of card games for children.
