Increased exposure to acute thermal stress is associated with a non-linear increase in recombination frequency and an independent linear decrease in fitness in Drosophila

Abstract

Background: Meiotic recombination rate has long been known to be phenotypically plastic. How plastic recombination evolves and is maintained remains controversial; though a leading model for the evolution of plastic recombination rests on the tenet that organismal fitness and recombination frequency are negatively correlated. Motivated by the mounting evidence that meiotic recombination frequencies increase in response to stress, here we test for a negative correlation between fitness and recombination frequency. Specifically, the fitness-associated recombination model (FAR) predicts that if stress increases meiotic recombination frequency, then increasing exposure to stressful conditions will yield an increasing magnitude of the recombinational response, while concomitantly decreasing fitness.

Results: We use heat shock as a stressor to test this prediction in Drosophila melanogaster. We find that increased exposure to heat shock conditions is associated with a non-linear increase in meiotic recombination frequency. We also find an independent effect of heat shock on organismal fitness, with fitness decreasing with increased duration of thermal stress.

Conclusions: Our results thus support the foundation of the FAR model for the evolution of plastic recombination. Our data also suggest that modulating recombination frequency is one mechanism by which organisms can rapidly respond to environmental cues and confer increased adaptive potential to their offspring.

Fig. 1

Schematic representation of the two-step crossing scheme using ebony (e) and rough (ro). Females used in each cross are shown on the left, males on the right. Boxed backcross 1 progeny (BC1) correspond to the two recombinant genotypes that can be visually identified using our screen

Fig. 2

Recombination frequency increases quadratically across exposure time. Each point represents a replicate. The gray line is the regression fitting time (P = 3.6 × 10⁻⁸) and time*time (P = 0.021). For the purpose of illustration, recombination frequency estimates displayed here were adjusted for batch effects. Note that the y-axis does not start at 0

Fig. 3

Number of offspring decreases across with increased exposure to heat shock conditions. The gray line is the regression fitting time (P = 2.4 × 10⁻⁴). For the purpose of illustration, values displayed here were adjusted for batch effects

Fig. 4

No correlation between recombination rate and number of offspring within time points. Open circles are individual replicates; filled circles are the mean values within time points. Random scatter is indicative of no correlation. In series A, the Y-axis is the recombination rate, where the increase with increased exposure can be seen by following the mean observations across time. In series B the axes are reversed so that number of offspring is along the Y-axis, and the decline with increased exposure time is apparent. Although recombination rate and number of offspring are negatively correlated overall, exposure time appears to be the effect driving changes in both factors independently, rather than recombination rate affecting number of offspring or vice versa