It's about time: the evidence for host plant-mediated selection in the apple maggot fly, Rhagoletis pomonella, and its implications for fitness trade-offs in phytophagous insects

The apple maggot fly, Rhagoletis pomonella, Walsh (Diptera: Tephritidae), p rovides a unique opportunity to address the issue of host-related fitness t rade-offs for phytophagous insects. Rhagoletis pomonella has been controver sial since the 1860's when Benjamin Walsh cited the fly's shift from hawtho rn (Crataegus spp.) to apple (Malus pumila) as an example of an incipient s ympatric speciation event. Allozyme and mark-release-recapture studies have subsequently confirmed the status of apple and hawthorn flies as partially reproductively isolated and genetically differentiated 'host races', the h ypothesized initial stage in sympatric divergence. Here, we review the ecol ogical and genetic evidence for host-plant mediated selection in R. pomonel la. We reach the following three major conclusions: First, although develop mental timing is not everything, it is a good deal of the story. Difference s in the fruiting phenologies of apple and hawthorn trees exert different s election pressures on the diapause and eclosion time characteristics of the host races. In particular, the similar to 3-week earlier mean fruiting phe nology of apples in eastern North America appears to select for a slower ra te of metabolism or deeper pupal diapause in apple than hawthorn flies. Sec ond, host-related fitness trade-offs for R. pomonella may not be due to dis ruptive selection affecting any one specific life-history stage. Rather, it is the sum total of directional selection pressures acting across differen t life-stages that generates divergent selection on apple and hawthorn flie s. For example, selection favors the alleles Me 100, Acon-2 95 and Mpi 37 ( or linked genes) in the larval stage in both host races. However, these sam e alleles are disfavored in the pupal stage to follow, where they correlate with early adult eclosion, and by inference premature diapause termination . Because apple trees fruit an average of similar to 3 weeks earlier than h awthorn trees, this counter-balancing selection is stronger on apple-fly pu pae. The net result is that the balance of selective forces is different be tween apple and hawthorn flies, helping to maintain the genetic integrity o f the host races in sympatry in the face of gene flow. Finally, natural R. pomonella populations harbor a good deal of genetic variation for developme nt-related traits. This variation allows fly populations to rapidly respond to temporal vagaries in local environmental conditions across years, as we ll as to broad-scale geographic differences that exist across the range of the species. Perhaps most importantly, this variation gives R. pomonella th e flexibility to explore and adapt to novel plants. Taken together, our res ults underscore how difficult it can be to document host plant-related fitn ess trade-offs for phytophagous insects due to the need to consider details of the entire life-cycle of a phytophagous insect. Our findings also show how reproductive isolation can arise as a by-product of host-associated ada ptation in insects, a central theme for models of sympatric speciation via host shifts.