Use of Genetic Technology to Understand Ecological and Behavioural Strategies of Bactrocera cacuminata (Hering)

Abstract

Many of Bactrocera are of particular concern throughout much of Asia and the Pacific, where they constitute a significant threat to agricultural production. Bactrocera cacuminata is a native and non-pest species in Australia, which can be used as a model species for studies of pest Bactrocera flies. Over the past four to five decades, knowledge of the ecology and biology of Bactrocera has been established based mostly on laboratory and semi field conditions. The behavioural strategies of Bactrocera have been broadly hypothesised as (1) females mate mainly only once; (2) adults emerging from a fruit will be full sibs, i.e. members of the same family; (3) there is genetic differentiation between regions but not between sites within regions and there is an overall pattern of IBD. Understanding aspects of the behaviour of Bactrocera flies is important for providing a context for other avenues of investigation including studies on pest management. The objective of this study was to use molecular techniques to test the hypotheses above in wild B. cacuminata. Specifically the aims were: 1) to estimate the level of polyandry, sperm utilization and sperm selection by analysing offspring genotypes from wild-caught females, 2) to estimate patterns of oviposition and larval development by analysing genotypes of flies emerging from wild fruit, 3) to estimate patterns of dispersal between populations on different spatial scales, within/between region(s). Estimates of gene flow (derived from hierarchical population genetic variation analyses) were used to infer patterns of dispersal. For this study, six polymorphic microsatellite loci and mtDNA gene, ND4 were developed. The microsatellites were isolated from enriched genomic libraries constructed using a biotin/streptavidin capture protocol. Allele number varied between three and nine; the expected heterozygosity ranged between 0.29 and 0.81. No significant deviations from Hardy–Weinberg equilibrium or linkage disequilibrium were found. The ND4 comprised 668 characters including 22 (3.3%) that were variable and 17 (2.5%) that were parsimony informative (18% in the 1st codon position, 82% in 3rd ). The fragment was free of ambiguities, stop codons and indels. Vary low levels of variability were found at ND4 and a number of other mtDNA genes in B. cacuminata. Female B. cacuminata was hypothesized matting only once and offspring have the same father. The level of polyandry, sperm utilization and kinship among flies were examined in a Brisbane wild population using five polymorphic microsatellite loci described above, plus an additional two loci developed for B. musae. Four hundred and twenty offspring from 22 wild-caught gravid females were genotyped to determine the number of males siring each brood and paternity skew, using the programs Gerud and Scare. The result showed that 22.7% of females produced offspring sired by at least two males. The mean number of mates per female was 1.72. Paternal contributions of double-sired broods were skewed with the most successful male having sired between 76.9% and 87.5% of the offspring. The polyandry and multiple paternity in B. cacuminata was against the hypotheses. The power of the paternity analysis showed that one sire was detected in 100% of simulations and 96.4% for two sires. These results have implications for a sterile insect technique (SIT), because the level of remating identified would indicate that wild females could mate with one or more resident fertile males, thus reducing the effectiveness of the technique...