Impact of the traditional mass rearing process on genetic variation in Aphytis melinus, revealed by double digest RAD sequencing”.

Aphytis melinus (Hymenoptera: Aphelinidae) is an important parasitoid of the California red scale, Aonidiella aurantii. This biocontrol agent has been commercially mass reared for nearly 50 years following its introduction into California in the late 1950s. Until the mid-1980s, quality control of natural enemies had not even been considered. Today there are quality control guidelines for some natural enemies, but testing is rare and there are no criteria specifically for A. melinus. Several studies have shown that biological control agents suffer from bottlenecks in population size, with subsequent loss of genetic variation. Furthermore, inbreeding and laboratory adaptation can negatively affect the fitness potential of biocontrol agents in the field. We used double digest Restriction-site Associated DNA sequencing (ddRADseq) to screen the genetic diversity of inbred and natural populations of A. melinus. Parasitoids were sampled from commercial insectaries in California, Spain and Australia; from a laboratory colony maintained for almost two decades and initiated with material collected in China; from five established Californian field populations; and from natural populations in Pakistan, which represents part of the native range of A. melinus. Fifty females or an equivalent mixture of both sexes per population were pooled, and each population was uniquely barcoded. Paired-end ddRADseq libraries were sequenced on an Illumina MiSeq. Raw Illumina reads were demultiplexed and processed using STACKS, a software pipeline that identifies SNPs and calculates population genetics indices (F statistics). We used SNP data derived from both read pairs analyzed separately to: perform Bayesian clustering in the program STRUCTURE; identify outlier loci under natural selection in the program BAYESCAN; and, infer phylogenetic trees using standard searching criteria (e.g., Maximum Likelihood). Both datasets suggested the presence of three clusters across our sampling: (1) Old World samples from the native range of A. melinus, (2) California/Australia field-collected samples and (3) samples obtained from insectaries. We examined genetic diversity across the sampled populations and found that samples obtained from insectaries had, in general, far fewer private alleles than any of the field-collected samples. These findings suggest that current protocols used in the mass rearing of parasitoids have led to reduced genetic diversity relative to other A. melinus populations. Interestingly, this reduction in diversity is not just compared to populations from the species’ native range, but also to those populations that have been putatively established from insectary stocks. We suspect that these patterns are the result of population bottlenecks and/or laboratory adaptation. Thus, our findings may have direct implications for the future management of this agriculturally important biocontrol agent.